Ozone is an important atmospheric species. In the troposphere, ozone can affect the atmospheric environment and human health. Moreover, it is a primary greenhouse gas (GHG). By absorbing longwave radiation, tropospheric ozone can alter the radiation balance and heat the atmosphere. Tropospheric ozone mainly originates from downward stratospheric ozone transport and the photochemical reaction of its precursors (nitrogen oxide, methane, carbon monoxide, and non-methane volatile organic compounds). Differing from other well-mixed GHGs, the distribution of tropospheric ozone is highly temporally and spatially dependent due to its short life of several weeks.

In China, there is significant seasonal variation in tropospheric ozone, with a maximum in summer and minimum in winter (Ziemke et al., 2011; Chen et al., 2015), which could be related to strong photochemical production and the high tropopause height in summer. Several studies based on observations from satellites, airplanes, and surface sites (Beig and Singh, 2007; Ding et al., 2008; Wang et al., 2009, 2012) found significant increases of tropospheric ozone in some areas of China, which could result from the increase in ozone precursors during the past several decades. Therefore, the climatic effects of tropospheric ozone may increase in the future and are worth studying.

The radiative effects of tropospheric ozone have drawn widespread attention. For example, Skeie et al. (2011) estimated a global radiative forcing of 0.44 W m^–2 in 2010 induced by increasing tropospheric ozone since 1750, corresponding to 24% of carbon dioxide radiative forcing. Meanwhile, Stevenson et al. (2013) calculated a tropospheric ozone radiative forcing of 0.41 W m^–2 for the period of 1750–2010 with 17 atmospheric chemistry models. Søvde et al. (2011) calculated a radiative forcing of 0.33 W m^–2 from ozone variation during the industrial era. Shindell et al. (2006) simulated tropospheric ozone from 1890 to 1990, and found that tropospheric ozone resulted in enhanced warming ( > 0.5°C) during the boreal summer in the polluted northern continental regions. In addition, Shindell et al. (2012) found that ozone and particularly aerosols resulted in significant precipitation changes over East and South Asia as well as the Sahel region in historical simulations. Chang et al. (2009) conducted research on the climate responses to the changes in long-lived GHGs, aerosols, and tropospheric ozone during 1951–2000, and found that ozone altered the average surface air temperature from 1971 to 2000 in eastern China by 0.43°C and changed the average precipitation by 0.08 mm day^–1. Xie et al. (2016) estimated an effective radiative forcing of 0.46 W m^–2 induced by changes in tropospheric ozone concentrations between 1850 and 2013. The resulting changes in the global annual mean surface temperature and precipitation were 0.36°C and 0.02 mm day^–1, respectively.

Overall, most studies have focused on radiative forcing derived from tropospheric ozone and the consequent impact on surface air temperature, and few studies have discussed the effects of tropospheric ozone on the monsoon circulation, precipitation, and feedback mechanism. Considering the increasing trend and clear seasonal changes in tropospheric ozone over eastern Asia, it is important to quantify and analyze its effects on regional climate to better understand tropospheric ozone–monsoon feedback. Therefore, in this study, the regional climate model RegCM4 was used to evaluate the effects of tropospheric ozone on the East Asian monsoon climate in summer and to analyze the possible influencing mechanisms.

A regional climate model (RegCM4) established by the International Centre for Theoretical Physics and reported by Giorgi et al. (2012) was employed to investigate the climatic effects of tropospheric ozone. This model includes a chemistry module (Solmon et al., 2006; Zakey et al., 2006, 2008; Shalaby et al., 2012) and can be used to study the spatial and temporal distribution of various air pollutants and their climatic effects (e.g., Solmon et al., 2008; Zhou et al., 2013; Li et al., 2016).

In this modeling study, the emission inventory of air pollutants and GHGs from the International Institute for Applied Systems Analysis was used (Höglund-Isaksson, 2012; Amann et al., 2013; Klimont et al., 2013; Stohl et al., 2013, 2015), which includes anthropogenic sources and open burning of agricultural residue. Regarding gas-phase chemical boundary species, monthly average concentrations as a climatological representative derived from the Model for OZone And Related chemical Tracers (MOZART) were employed. A newly implemented Rapid Radiation Transfer Model (RRTM), as reported by Mlawer et al. (1997), was used to represent the radiation process. Moreover, the Holtslag boundary layer scheme (Holtslag et al., 1990), large-scale precipitation scheme (Pal et al., 2000), and Emanuel cumulus convection parameterization scheme (Emanuel, 1991; Emanuel and Živković-Rothman, 1999) were used in the model. The modeling domain covered China and its neighboring regions with 18 vertical layers and a 60-km horizontal resolution. The pressure value at the top of the model was 50 hPa. This model was driven by meteorological data from the ERA-Interim (Dee et al., 2011). Sea surface temperature data with a 1° × 1° resolution (Reynolds et al., 2002) were obtained from the NOAA. To estimate the radiative effect from tropospheric ozone, the climatology of tropopause pressure from the NCEP/NCAR reanalysis (Kalnay et al., 1996) was used. The simulation time included the months from May to August in 2001–10, with May used as the spin-up time.

Two groups of numerical experiments were designed to investigate the climatic effects of increased tropospheric ozone. For Experiment 1 (Exp1), the climatological data of preindustrial (1850) tropospheric ozone from the SPARC (Stratosphere–troposphere Processes And their Role in Climate) ozone database (Cionni et al., 2011) were employed in the radiation module. For Experiment 2 (Exp2), the real-time calculated ozone concentration from the chemistry module was transferred to the radiation module. Both Exp1 and Exp2 considered the radiative effects of tropospheric ozone. By comparing Exp1 and Exp2, the impact of increased tropospheric ozone on the regional climate of East Asia could be evaluated. In Exp2, the instantaneous radiative forcing, holding state variables (e.g., water vapor, tropospheric temperature, and clouds) fixed at the unperturbed values, was calculated by calling the radiative subroutine twice with the concentrations of preindustrial and present-day tropospheric ozone.

The observed column concentration of tropospheric ozone for model validation in the following section was obtained from the NASA’s Goddard Space Flight Center. These data, with a resolution of 1.25° × 1° (valid from October 2004 onward), were derived from the ozone measurements from the Ozone Monitoring Instrument (OMI) and Microwave Limb Sounder (MLS) onboard on the Aura satellite using the tropospheric ozone residual method (Ziemke et al., 2006, 2011). These data were verified with ozonesonde data and showed good reliability.

Figure 1 shows the average column concentration of tropospheric ozone from the observations and simulations in summer (JJA) from 2005 to 2010. Ozone was largely distributed in Central China, North China, East China, and the Sichuan basin, with a maximum of about 55 Dobson Units (DU). This could have been due to the heavy traffic and highly developed industries in East and North China, which produce large amounts of ozone precursors. In addition, the summer monsoon may be a crucial influencing factor of the distribution of ozone. For example, southerly winds can blow air pollutants to northern China. This pattern showed similar distribution characteristics with previous simulation studies (e.g., Wang et al., 2005; Hou et al., 2016). Compared to the observations, the simulation captured the distribution of ozone well, although there were overestimates in most parts of China and underestimates in some parts of North China and Northwest China. The column concentration differences could be related to the lack of some natural emission sources such as biogenic emissions and imperfections in the physics and chemistry modules used in the model.

Figure 1 Summer (June–August) mean column concentration (DU) of tropospheric ozone from (a) observations and (b) simulations (present-day) (2005–10). The observation data are from Aura OMI/MLS (Ziemke et al., 2006).

Figure options

Figure 2 mainly describes the clear-sky shortwave and longwave radiative forcings and all-sky shortwave and longwave radiative forcings at the tropopause from increased tropospheric ozone since pre-industrial times. Ultraviolet radiation (wavelength < 0.3 μm) is mainly absorbed by the stratospheric ozone; therefore, the clear-sky shortwave radiative forcing was small, with high values in the east part of China. Meanwhile, clear-sky longwave radiative forcing was large in North China, East China, and Central China, with a maximum of approximately 0.8 W m ^–2. The distributions of shortwave and longwave radiative forcings showed good correlations with the distribution of the tropospheric ozone column concentration.

The all-sky shortwave radiation forcing was much larger than the clear-sky shortwave radiation forcing, whereas the all-sky longwave radiation forcing was smaller than the clear-sky longwave radiation forcing. These differences may be attributed to the impact of clouds. Clouds can increase the albedo effect and reflect shortwave radiation, thereby making ozone absorb more shortwave radiation. Meanwhile, clouds can absorb longwave radiation and make the ozone over the cloud absorb less longwave radiation from the surface.

All-sky shortwave radiation forcing showed a different distribution from clear-sky shortwave radiation forcing, with high values in South China and Northeast China, which could have been caused by more clouds in these areas. Meanwhile, the distribution of all-sky longwave radiation forcing was similar to that of clear-sky longwave radiation forcing.

Table 1 shows the average shortwave, longwave, and total radiative forcings due to increased tropospheric ozone over southern China (22°–32°N, 108°–122°E), northern China (32°–42°N, 108°–122°E), eastern China (22°–42°N, 108°–122°E), and the whole modeling domain. The results from several other studies are also listed in Table 1. The regional average values over the modeling domain for clear-sky shortwave and longwave radiative forcings were 0.14 and 0.54 W m^-2, respectively, similar to the radiative forcing over China from other studies, and was higher than the estimated global radiative forcing of 0.40 ± 0.20 W m^-2 proposed by the Intergovernmental Panel on Climate Change (IPCC) (2013) and several other global results. This is indicative of significant heating effects of ozone over East Asia, particularly over eastern China, in summer.

Figure 2 Summer (June–August) mean clear-sky (a) shortwave and (b) longwave radiative forcings, and all-sky (c) shortwave and (d) longwave radiative forcings (W m–2) at the tropopause due to tropospheric ozone. The red contour lines represent the total cloud amount (%).

Figure options

Figure 3 illustrates the net shortwave and longwave radiative flux (downward) changes and total cloud amount changes at the tropopause due to tropospheric ozone. The radiative flux changes were much larger than the radiative forcing due to the climate response. There were several corresponding relationships between the distributions of radiative flux change and all-sky radiative forcing. Moreover, the change in amplitude in the shortwave radiative flux was larger than that of the longwave radiative flux, which could have been associated with the change in the cloud amount (Fig. 3c) due to climatic adjustment. Meanwhile, the change in the cloud amount could have been due to the change in radiation balance and resulting circulation change (see discussion below).

Figure 3 Summer (June–August) mean net (a) shortwave and (b) longwave radiative flux changes (W m–2) at the tropopause, and (c) total cloud amount change (%) due to tropospheric ozone. Dotted areas denote results passing the t-test at the 90% confidence level (hereinafter inclusive).

Figure options

Figure 4 presents the changes in the surface air temperature caused by the tropospheric ozone. The tropospheric ozone resulted in an increase in the surface air temperature over most parts of East Asia, with a maximum of 0.2 K in East China, North China, and Northwest China. Combined with Fig. 3, we found that not only the absorption of longwave radiation by ozone, but also the cloud amount anomaly and corresponding shortwave radiation anomaly, had important effects on the surface air temperature change in China.

Figure 4 Summer (June–August) mean surface air temperature change (K) due to tropospheric ozone.

Figure options

Figure 5 shows the geopotential height field and wind vector at different isobaric surfaces and their changes induced by tropospheric ozone. Southerly wind prevailed in the lower troposphere over eastern China during summer, carrying large amounts of water vapor from the south ocean. There were a strong cyclonic wind field change and a geopotential height field anomaly over Northeast China, the Sea of Japan, and the Yellow Sea, which extended to Central China. This cyclonic wind field change became weaker with increasing height. Consequently, southerly wind decreased in northern China and increased in southern China in the lower troposphere.

Figure 6 illustrates the zonally averaged (108°–122°E) air temperature (shaded) and vertical wind field (stream) as well as their changes due to tropospheric ozone. Ozone caused an increase in air temperature in the whole atmosphere. The notable positive air temperature anomaly near the surface may have been caused by ozone from increased anthropogenic emissions. Meanwhile, temperature anomalies in the upper troposphere may have resulted from the latent heat release related to an enhancement in upward motion. There was a marked increase in air temperature over the area around 30°N, which led to a decrease in air density and strengthening of upward motion. The consequent convergence near the ground may have had a contributing role in enhancing the southerly wind over the region south of 30°N. Overall, the meridional land–sea thermal contrast increased due to significant heating over land, and thus, the meridional wind was strengthened to the south of the heating center.

Figure 5 Summer (June–August) (a, b, c) mean geopotential height field (shaded; gpm) and wind vector (arrow; m s–1), and (d, e, f) their changes at (a, d) 925 hPa, (b, e) 850 hPa, and (c, f) 500 hPa due to tropospheric ozone.

Figure options

Figure 6 Summer (June–August) (a) mean zonally averaged (108°–122°E) vertical wind stream and air temperature (shaded; K), and (b) their changes due to tropospheric ozone.

Figure options

Figure 7 illustrates the changes in precipitation and zonally averaged (108°–122°E) specific humidity due to increased tropospheric ozone in summer. Tropospheric ozone caused an increase in precipitation in many parts of China, with a maximum of 3 mm day^–1 in the middle and lower reaches of the Yangtze River. This could have been related to the enhanced ascending motion to the south of 34°N (Fig. 6). In addition, the southerly wind anomaly could carry more water vapor into the area (see Fig. 7b), which would be beneficial to the formation of precipitation. Meanwhile, the precipitation decrease in North China could be due to the weakening of the ascending motion and negative southerly wind anomaly. The precipitation change in pattern indicates that the circulation change induced by tropospheric ozone has an important influence on precipitation.

Figure 7 (a) Changes in summer (June–August) mean precipitation (mm day–1) and (b) zonally averaged (108°–122°E) specific humidity (g kg–1) due to tropospheric ozone.

Figure options

Table 2 presents the regionally averaged statistical results of the climatic effects due to the tropospheric ozone over different regions. Tropospheric ozone led to an increase in the net absorbed solar flux, an increase in the net longwave radiative flux (downward) at the top of the atmosphere, and an increase in surface air temperature over East Asia. In addition, increased tropospheric ozone enhanced the monsoon circulation over southern China and corresponding precipitation. Meanwhile, the effects showed the opposite trend over northern China. The climatic effects over eastern China were more significant than those over other areas of the modeling domain that corresponded to the distribution of tropospheric ozone. Compared to previous studies on the influence of the tropospheric ozone in China (Wu et al., 2003; Wang et al., 2004; Chang et al., 2009), there were several differences in the distribution and magnitude of the temperature change. Nevertheless, all of the simulations indicated that the tropospheric ozone resulted in an increase in the average surface air temperature over China. The differences between our work and other studies could be due to differences in the models and experimental designs. In addition, the absence of feedback at a larger scale that cannot be included in regional models may explain these differences.

Notably, the precipitation change due to increased tropospheric ozone showed a “southern flood and northern drought” pattern, similar to the influence of black carbon aerosols proposed by Menon et al. (2002). This could be because tropospheric ozone has similar radiative properties as black carbon. Both tropospheric ozone and black carbon heat the air over southern China, changing the meridional circulation and hydrologic cycle. Several studies (e.g., Wu et al., 2013; Song et al., 2014; Li et al., 2015; Zhang and Li, 2016) have examined the contribution of increased GHGs and anthropogenic aerosols on the East Asian summer monsoon variation. GHGs and anthropogenic aerosols have been found to produce competing effects on monsoon circulation and rainfall over East Asia. Anthropogenic aerosols lead to reduced rainfall and weakened monsoon circulation, while GHGs produce the opposite effects. This study illustrates that tropospheric ozone generally generates an impact similar to that of GHGs on the East Asian summer monsoon, particularly in southern China. In view of the high concentrations over East Asia, it is necessary to consider the radiative effects of tropospheric ozone in studies on changes in the East Asian summer monsoon.

In this study, the tropospheric ozone distribution, radiative forcing, and climatic effects due to the increase in tropospheric ozone since the industrialization era in China during summer were investigated by using the regional climatic model RegCM4.

The column concentration of tropospheric ozone was large in East China, North China, Central China, and the Sichuan basin because of the distribution of ozone precursor emissions and the influence of the summer monsoon. Tropospheric ozone produced positive shortwave radiative forcing and positive longwave radiative forcing, which resulted in increases in the summer mean surface air temperature and precipitation over eastern Asia. The change in air temperature was significant in East China, Northwest China, and North China, which could be attributed to the absorption of longwave radiation by ozone and the negative cloud amount anomaly and corresponding positive shortwave radiation anomaly. The precipitation change was notable in the middle and lower reaches of the Yangtze River, where an anomaly of upward motion occurs. Enhanced southerly winds could carry more water vapor to this area, and would also be favorable to increasing precipitation. In addition, the enhancement of the monsoon circulation caused by tropospheric ozone strengthened both zonal wind and meridional wind at the lower troposphere over southern China in summer.

Overall, increased tropospheric ozone resulted in an increase in surface air temperature and an enlarged land–sea thermal contrast, thereby strengthening the summer monsoon and increasing precipitation. However, the feedbacks between the tropospheric ozone and regional climate are complex and have large uncertainties, which could result in a distribution of ozone somewhat inconsistent with the strength of the climatic effect.

This study yielded several preliminary conclusions regarding the climatic effects of increased tropospheric ozone in China in summer. Regardless, a number of questions remain unanswered. For example, the effects of aerosols on ozone via their effects on the photolysis rate and heterogeneous reactions have not been considered in the regional climate model. Moreover, the climatic effect caused by each ozone precursor emission has not been analyzed. These issues should be explored in the future research.