Effect of Methane Emission Increases in East Asia on Atmospheric Circulation and Ozone

Received 22 January 2015; revised 10 May 2015; accepted 4 June 2015

Methane (CH₄) is both an important greenhouse and chemically active gas. In the last glacial period (about 18 000 years ago), the concentration of CH₄ was 0.35 ppmv, but it reached 0.75 ppmv in the mid-19th century (e.g., Rasmussen and Khalil, 1984; Chappellaz et al., 1990). With the growth of human activities, the tropospheric CH₄ concentration reached 1.72 ppmv in 1995 (e.g., Lelieveld et al., 1998). (Zhang et al., 2001) showed that from the end of the 20th century to the beginning of the 21st century, CH₄ emissions increased by 45% during three decades. (Ramaswamy et al., 2001) pointed out that when the 1750 CH₄ mixing ratio of 0.72 ppmv increased to 1.77 ppmv in 2005, a radiative forcing of 0.48 0.05 W m^-2 was the result. CH₄ is the second most influential radiative-forcing greenhouse gas after CO₂, excluding water vapor.

After the signing of the Montreal Protocol, atmospheric ozone (O₃) depleting substances (ODSs) such as chlorides have been slowly reduced, while CH₄ may continue to increase significantly during the next 50 years. Satellite data show that in East Asia, with China's rapid economic growth, the concentration of pollutants has continued to grow in recent years. (Bergamaschi et al., 2013) found that from 2003 to 2008 CH₄ emissions caused by human activities grew fastest in China (18.6 Tg yr^-1) —— far faster than Brazil with the second fastest growth rate of 2.5 Tg yr^-1. (Zhang et al., 2011) analyzed Atmospheric Infrared Sounder satellite data and found that, compared with several major countries worldwide, the growth rate of CH₄ emissions has been significantly higher in China than other countries since 2007.

In addition to changing the energy balance of the Earth's atmosphere through the greenhouse effect, CH₄ emission increases also affect water vapor and the O₃ concentration in the stratosphere. Most CH₄ is oxidized in the troposphere by hydroxyl radicals (OH) (e.g., Levy, 1971), but about 7%-11% of CH₄ is transported into the stratosphere and oxidized (e.g., Born et al., 1990). CH₄ oxidation is an important source of stratospheric water vapor (e.g., Lelieveld and Crutzen, 1992; Fuglestvedt et al., 1996; Shindell et al., 2005). CH₄ emissions increase the O₃ concentration in the troposphere and stratosphere (e.g., Owens et al., 1982, 1985; Isaksen and Stordal, 1986; Wuebbles and Hayhoe, 2002). During the 20th century, the global background O₃ concentration rose by at least a factor of two, due mainly to increases in CH₄ and nitrogen oxides (NO_x) emissions (e.g., Marenco et al., 1994; Wang and Jacob, 1998).

Based on a series of model simulations, (Butchart and Scaife, 2001) found that mass flux transported from the tropical troposphere to stratosphere increases at a rate of 3% (10 yr)^-1 under conditions of increasing greenhouse gases, as supported by the Intergovernmental Panel on Climate Change (IPCC) emissions scenarios. Increasing greenhouse gases enhances wave activities in the lower stratosphere and accelerates the Brewer-Dobson (BD) circulation (e.g., Rind et al., 2001; Sigmond et al., 2004; Butchart et al., 2006). O₃, water vapor and other trace gases in the troposphere are transported into the stratosphere faster as the BD circulation strengthens (e.g., Austin et al., 2007). (Austin and Li, 2006) found that greenhouse gas and BD circulation increases jointly lead to increased CH₄ transport into the stratosphere, which is then oxidized into water vapor. The increase of stratospheric water vapor reduces the average age of the atmosphere in the stratosphere. The average age of the atmosphere indicates the transport time scales of substances leaving the troposphere and entering into the stratosphere (e.g., Hall and Plumb, 1994; Waugh and Hall, 2002).

The effects of a CH₄ concentration increase on atmospheric radiation and chemistry has been studied extensively (e.g., Shindell et al., 2005; Chen et al., 2006; Shi, 2006; Bi et al., 2007, 2008; Guo et al., 2008; Bi, 2009; Shi et al., 2009; Xie et al., 2013). (Gruzdev and Brasseur, 2005) used a two-dimensional model SOCRATES (Simulation of Chemistry, Radiation, and Transport of Environmentally Important Species, SOCRATES) and found that the stratosphere and middle atmosphere became cooler with an amplitude of less than 1 K when CH₄ was increased by 26%. (Dyominov and Zadorozhny, 2005) found CH₄ increases can cause an increase in stratospheric O₃ and lead to a warming in the stratosphere. However, the chemistry-radiation-dynamic interactions associated with CH₄ increases have not been well resolved in previous modeling studies, due to the limitation of model performance in the stratosphere. On the other hand, most previous studies have investigated the effects of global-scale CH₄ emission increases on climate and stratospheric O₃, but the effect of a CH₄ emissions increase in East Asia on global temperature and O₃ is also still a popular and controversial issue. (Wild and Palmer, 2008) used a global chemistry transport model to demonstrate that the spatial extent of O₃ production and loss in the troposphere changes very little despite large projected increases in precursor emissions. (Stevenson et al., 2006) used 26 atmospheric chemistry models to reveal that ensemble-mean changes in the tropospheric O₃ burden between 2000 and 2030 range significantly, but the effect of CH₄ increases on the tropospheric O₃ was not addressed. In this paper, we use a fully coupled chemistry-climate model to investigate the impact of a CH₄ emissions increase in East Asia on the stratospheric temperature, circulation and O₃, and tropospheric O₃. The results are compared with the impact of global CH₄ increases and a CH₄ emissions increase in North America, to assess the contribution of anthropogenic emissions in East Asia to climate change.

We used version 3 of the Whole Atmosphere Community Climate Model (WACCM3), which is a global chemistry-climate model with 66 vertical levels extending from the surface to 4.5× 10^-6 hPa (∼160 km). WACCM3 is based on the software framework of the National Center for Atmospheric Research's Community Atmosphere Model, version 3 (CAM3), and includes all of the physical parameterizations of that model. The governing equations, physical parameterizations and numerical algorithms used in CAM3 are documented by (Collins et al., 2004), and only the gravity wave drag and vertical diffusion parameterizations are modified in WACCM3. The WACCM3 chemistry module is derived from the three-dimensional chemical transport model MOZART (Model for Ozone and Related Chemical Tracers) (e.g., Brasseur et al., 1998; Hauglustaine et al., 1998; Horowitz et al., 2003). This model resolves 51 neutral species, including all members of the O _x (oxides), NO _x, HO _x (hydrogen oxides), ClO _x (chlorine oxides) and BrO _x (bromine oxides) chemical families, along with tropospheric "source species" such as N₂O (nitrous oxide), H₂O, CH₄, chlorofluorocarbons (CFCs) and other halogenated compounds. WACCM3 performs well in simulating chemical and other atmospheric processes (e.g., Garcia et al., 2007; Eyring et al., 2010; Liu and Liu, 2009). Compared with observation, its simulations tend to be too warm in the Arctic winter, and produce cold temperatures and westerly winds that persist for too long in the Antarctic spring (e.g., Garcia et al., 2007). These biases influence the temperature and O₃ concentration in the polar regions, so the simulation results in the polar regions are not discussed in detail in this study.

Figure 1.  Latitude-height cross sections of temperature differences (units: K) in (a) E15, (b) E30, (c) E50, (d) A50 and (e) G15, relative to E0. Shaded regions show where the differences are statistically significant at the 90% confidence level, using the Student's t-test.

We conducted six simulations at a horizontal resolution of 4^°× 5^° with interactive chemistry, including one control experiment and five sensitivity experiments, listed in Table 1. The surface emissions of CO₂, CH₄, Cl (chlorine), Br (bromine) and 14 other ODSs used in the control run (E0) are 12-month climatologies derived from IPCC's A1B scenario (IPCC, 2007) averaged over the period from 1979 to 2006. The A1 scenario means a future world of very rapid economic growth, a global population that peaks mid-century and declines thereafter, and the rapid introduction of new and more efficient technologies. Under the A1B scenario, technological changes occur in the energy system in a balanced manner across all sources. It has also been widely used in previous chemistry-climate model simulations in the literature to represent past and future ODS variations. The SST and sea ice fields used in the model are 12-month climatologies derived from (Rayner et al., 2003). According to the IPCC 2007 A1B scenario, globally averaged CH₄ emissions increased by 15% in 2006, compared to 1979, and are expected to increase by 50% in 2050, meaning CH₄ emissions will increase observably in the future. In order to study the effect of different CH₄ emission increases in East Asia, in experiments E15, E30 and E50 we forced the surface CH₄ emissions in East Asia to increase by 15%, 30% and 50%, respectively, compared to experiment E0 (the increments were 7.5, 15 and 25 Tg yr^-1, approximately). East Asia was taken to range over the area (0^°-90^°N, 60^°-120^°E). In order to study the effect of CH₄ emission increases in different areas, in experiment A50 the surface CH₄ emissions in North America were increased by 50%, the increment (25 Tg yr^-1, approximately) and area of which were equal to E50 and the range of North America was taken as (0^°-90^°N, 60^°-120^°W). In experiment G15 we forced global surface CH₄ emissions to increase by 15% (increment: 90 Tg yr^-1, approximately). The simulations were run for 50 years (from 2000 to 2050), with the first 10 years used as model spin-up, and the remaining 40 years (from 2010 to 2050) of model outputs used for analysis.

Figure 2.  Latitude-height cross sections of (a, c, e) EP flux (arrows), where shading represents the differences of the average zonal westerly winds (units: m s^-1), with positive change indicating strengthening westerlies and negative change indicating weakening westerlies; and (b, d, f) BD circulation differences (arrows): (a, b) E50 relative to E0; (c, d) A50 relative to E0; (e, f) G15 relative to E0.

Figure 1 shows the stratospheric temperature differences of E15, E30, E50, A50 and G15 relative to that in E0. When CH₄ emissions increase by 15% in East Asia, the temperature increases in the lower stratosphere in the Southern Hemisphere, but the temperature differences are not statistically significant (Fig. 1a). When CH₄ emissions increase by 30% in East Asia, there is a weaker temperature decrement in the lower stratosphere of the Southern Hemisphere compared with E15, but the temperature increases in the upper stratosphere (Fig. 1b). The temperature difference is still not statistically significant in E30 either. When compared with the temperature differences induced by CH₄ emission increases of 15% and 30%, a significant climate effect appears when CH₄ increases in East Asia by 50%, causing the middle stratosphere to cool significantly with a maximum cooling of 0.15 K at 30^°N at 50 hPa (Fig. 1c). Therefore, we base our conclusions on the results of E50 when studying the climate effect of CH₄ emission increases in East Asia in the following analysis. Figure 1 also indicates that a 50% CH₄ increase in North America causes a cooling in the middle stratosphere, with a maximum of 0.15 K at 30^°N at 50 hPa. When global CH₄ emissions increase by 15%, the amplitude of the stratosphere maximum cooling is similar to that in E50 at 20^°S at 60 hPa, but the range of cooling is larger than that in E50. That is to say, the greenhouse effect is much larger in experiment G15 than E50.

Figure 2 shows the differences of the Eliassen-Palm (EP) flux and BD circulation of experiments E50, A50 and G15 compared with experiment E0. The BD circulation is a global-scale cell in the stratosphere in which air rises in the tropics and then moves polewards and downwards, mostly in the winter hemisphere. The BD circulation describes Lagrangian-mean transport and the transformed Eulerian mean residual velocities (v^*, w^*) (e.g., Andrews and McIntyre, 1976, 1978), which approximate the mean meridional mass transport for seasonally averaged conditions (e.g., Holton, 1990). When CH₄ emissions increase, the stratospheric westerly weakens in the Southern Hemisphere midlatitudes and the mid-stratospheric westerly in the Northern Hemisphere strengthens. Compared with E0, the maximum westerly increment appears in the Northern Hemisphere in E50 and A50, and in the Southern Hemisphere in G15: the location is consistent with Fig. 1, where the temperature gradient is evident. When the extratropical temperature gradient increases, the westerly and wave activity increase (e.g., Eichelberger and Hartmann, 2005; Olsen et al., 2007; Garcia and Randel, 2008). As shown in Fig. 2, as the wave force becomes stronger in the Northern Hemisphere midlatitudes, the vertical velocity of BD circulation speeds up. So, the changes of the mean zonal wind, EP flux and BD circulation are consistent, as shown in Fig. 2. (Fomichev et al., 2007) found that the EP flux increases when the westerly is stronger at midlatitudes and the wave flux change in the range 20^°-40^°N is responsible for the tropical upwelling change. (Garcia and Randel, 2008) found that the maximum increase of EP flux appeared in the tropics during 1950-2003, and in the midlatitudes during 1980-2050, causing the BD circulation to change accordingly. It is worth noting that the vertical velocity of EP flux at 100 hPa in the Northern Hemisphere midlatitudes changes most significantly in all the sensitivity experiments, and the vertical velocity of BD circulation also increases significantly south of the equator in A50; but, in E50 and G15, the vertical velocity of BD circulation reduces south of the equator at 100 hPa. This is further analyzed later in the paper.

Using a series of models (including WACCM), (Butchart et al., 2006) found that BD circulation upwelling regions extend roughly 30^° either side of the equator, though there are variations in the width of about 10^° between the models. In our simulations, the BD circulation rises between 22^°S and 30^°N under the SST and greenhouse gas concentrations used. So, the region between 22^°S and 30^°N is defined as the BD circulation rising zone. (Lin and Fu, 2013) found that the acceleration rate of the BD circulation is much larger in the lower stratosphere than upper stratosphere as greenhouse gas concentrations increase. As previous studies have used the upward mass flux across 100 hPa to infer the strength of the BD circulation (e.g., Garcia and Randel, 2008; Garny et al., 2011), Table 2 shows the changes of upward mass flux of E50, A50 and G15 at different altitudes. The increment of upward mass flux is larger at 100 hPa than at 70 hPa and 30 hPa, i.e., when CH₄ emissions increase (in East Asia, North America and on the global scale) the increment of upward mass flux and the acceleration of the BD circulation is larger in the lower than in the upper stratosphere. It is worth noting that the change of upward mass flux is largest in experiment A50; so, compared with a 50% CH₄ emissions increase in East Asia, there is a larger effect of the CH₄ emissions increase in North America on the EP flux, BD circulation and mass flux.

In order to study the BD circulation vertical velocity difference as CH₄ emissions increase by different concentrations and in different areas, Fig. 3 further shows the difference of the BD circulation vertical velocity between sensitivity experiments and the control experiment E0. When CH₄ emissions increase in East Asia and globally, the difference of w^* is rather small between 22^°S and 10^°S, while in the equatorial region (10^°S-10^°N) w^* reduces and w^* is stronger between 10^°N and 30^°N at 100 hPa. When CH₄ emissions increase in North America, the difference of w^* between 22^°S and 10^°S is similar to that in other experiments, but it is very different between 10^°S-30^°N, as in the equatorial region (10^°S-10^°N) w^* becomes stronger and between 20^°-30^°N w^* is weaker. In the following analysis, the region between 22^°S and 30^°N is divided into three parts: 10^°-30^°N, 10^°S-10^°N and 10^°-22^°S.

Figure 3.  Difference of BD circulation vertical velocity w^* (units: -Pa s^-1) at 100 hPa. Lines represent the difference of E15 (black), E30 (blue), E50 (green), A50 (orange) and G15 (red) relative to E0.

CH₄ emission increases accelerate the BD circulation, which can transport more air mass from the troposphere into the stratosphere. Table 3 shows the differences of upward mass flux at 100 hPa in experiments E15, E30, E50, A50 and G15, relative to that in experiment E0. Note that the upward mass flux between 10^°S and 10^°N decreases in all experiments except A50, i.e., E15, E30, E50 and G15, compared to that in experiment E0. The changes of upward mass flux are consistent with BD circulation vertical velocity variations.

Figure 4 shows the seasonal variations of the BD circulation vertical velocity. It is apparent that the vertical velocity variations caused by different levels of increase in CH₄ surface emissions are quite different, although the vertical velocity reduces in January and February in all the sensitivity experiments between 10^°S and 10^°N, and in this range the BD circulation ascending branch becomes stronger during March and June in E15. This change is not obvious in E30, and the vertical velocity reduces during May and August in E50. Table 3 shows that between 22^°S and 10^°S the upward mass flux decreases in E15 and E30, but increases in E50. When CH₄ emissions increase in East Asia the increment of upward mass flux between 10^°N and 30^°N is largest in E30. Above all, the change of upward mass flux at 100 hPa is nonlinear as the CH₄ emissions gradually increase in East Asia in E15, E30 and E50. In A50 the vertical velocity decreases slightly in January and February between 10^°S and 10^°N and increases obviously between 10^°N and 20^°N from January to May. The combined effect is that the upward mass flux increment at 100 hPa is largest in A50. There is no obvious change of mass flux between 10^°S and 10^°N in G15, but the mass flux increase is remarkable between 10^°N and 30^°N, as indicated by the noteworthy vertical velocity increase in January and February and the slight increase in other months. Therefore, there is also a large increase of the total mass flux in G15 between 22^°S and 30^°N.

Figure 4.  Seasonal variation of BD circulation vertical velocity w^* (-Pa s^-1) at 100 hPa: the difference of (a) E15, (b) E30, (c) E50, (d) A50 and (e) G15, relative to E0.

Figure 5.  Latitude-height cross sections of O₃ differences (units: ppbv) in (a) E50, (b) A50 and (c) G15, relative to E0. Shaded regions show where the differences are statistically significant at the 90% confidence level, using the Student's t-test.

Figure 6.  Latitude-height cross sections (a-c) of OH differences (units: %) and (d-f) O₃ differences (units: %) of (a, d) E50, (b, e) A50 and (c, f) G15, compared with E0. Shaded regions show where the differences are statistically significant at the 90% confidence level, using the Student's t-test.

Previous studies have shown that the tropospheric and stratospheric O₃ increases as CH₄ emissions increase (e.g., Owens et al., 1982, 1985; Isaksen and Stordal, 1986; Wuebbles and Hayhoe, 2002). Figure 5 shows the difference of stratospheric O₃ in E50, A50 and G15. In the middle stratosphere, CH₄ reacts with active chlorine (CH₄ + Cl \(\to\) CH₃ + HCl) and decreases catalytic O₃ loss caused by Cl radicals; this is the main cause of O₃ increase in the middle stratosphere.

When CH₄ emissions increase by 50% in East Asia (E50) and 15% on the global scale (G15), the maximum increase of stratospheric O₃ is 45 ppbv and 60 ppbv, respectively. Therefore, a CH₄ emissions increase is beneficial for stratospheric O₃ recovery. It should be noted that although the increment of CH₄ emissions in East Asia and North America is the same, the stratospheric O₃ increase in A50 is larger, and the maximum increase reaches 60 ppbv. That is to say, when CH₄ emissions increase by the same amount in East Asia and North America, the effect on the stratospheric O₃ of the former is less than the latter.

Around 90% of atmospheric CH₄ is removed through chemical reactions in the atmosphere, and 10% is removed by soil oxidation. Figure 6 shows the OH and O₃ difference in the troposphere between each sensitivity experiment and experiment E0. The reaction between CH₄ and OH radicals (CH₄ + OH \(\to\) CH₃ + H₂O) is the main removal mechanism of CH₄ in the troposphere. In the stratosphere, CH₄ reacts with the OH and O¹(D) radical; however, the photolysis process of CH₄ can be ignored (e.g., Qin and Zhao, 2003). The photochemical reaction of the OH radical is important for oxidation and removal of numerous trace gases. In general, the concentration of the OH radical is highest in the tropics, and slightly higher in the Southern than the Northern Hemisphere [e.g., (Chameides and Davis, 1982)]. As shown in Figs. 6a and b, OH in the Northern Hemisphere significantly reduces as CH₄ oxidation by OH mainly occurs in the Northern Hemisphere when CH₄ emissions increase in East Asia and North America. CH₄ is oxidized by OH on the global scale in G15, so the tropospheric OH reduces uniformly in both hemispheres.

One quarter of the hydroxyl radicals react with CH₄, starting several oxidation chains that strongly affect the atmospheric concentrations of OH and O₃ (e.g., Crutzen, 1973). The availability of NO plays an important role in determining the oxidation pathways: \begin{eqnarray*} &&{\rm CH}_{4}+{\rm OH}\to {\rm CH}_{3}+{\rm H}_{2}{\rm O}\\ &&{\rm CH}_{3}+{\rm O}_{2}+{\rm M}\to {\rm CH}_{3}{\rm O}_{2}+{\rm M}\\ &&{\rm CH}_{3}{\rm O}_{2}+{\rm NO}\to{\rm CH}_{3}{\rm O}+{\rm NO}_{2}\\ &&{\rm CH}_{3}{\rm O}+{\rm O}_{2}\to {\rm CH}_{2}{\rm O}+{\rm HO}_{2}\\ &&{\rm HO}_{2}+{\rm NO}\to {\rm OH}+{\rm NO}_{2}\\ &&{\rm NO}_{2}+{\rm h}\nu\to{\rm NO}+{\rm O}(\le400\;{\rm nm}) \end{eqnarray*} \begin{eqnarray*} &&{\rm O}+{\rm O}_{2}+{\rm M}\to{\rm O}_{3}+{\rm M}\\ &&{\rm Net:} {\rm CH}_{4}+{\rm N}{\rm O}+{\rm 3O}_{2}\to {\rm CH}_{2}{\rm O}+{\rm H}_{2}{\rm O}+{\rm N}{\rm O}_{2}+{\rm O}_{3} \end{eqnarray*}

Figure 7.  Longitude-latitude cross sections of O₃ differences (units: %) at (a-c) 500 hPa and (d-f) 200 hPa in (a, d) E50, (b, e) A50 and (c, f) G15, compared with E0. Shaded regions show where the differences are statistically significant at the 90% confidence level, using the Student's t-test.

CH₄ oxidation by OH mainly occurs in the Northern Hemisphere troposphere when CH₄ emissions increase in East Asia and North America. This causes O₃ to increase significantly in the Northern Hemisphere troposphere. The concentration of CH₄ is higher at high latitudes than at low latitudes. With the same percentage increase of CH₄ in East Asia and North America, the increment of CH₄ at high latitudes is larger than at low latitudes. But, as shown in Figs. 6d and e, the maximum percentage increase of the tropospheric O₃ appears in the low-and midlatitudes. This is because CH₄ oxidation by OH mostly occurs in the tropics and the concentration of NO_x is largest in the midlatitudes (e.g., Fiore et al., 2008). It should be noted that OH reduces by 4%, and O₃ in the Northern Hemisphere midlatitudes increases by 4% in E50. There is a small variation in A50, with OH reducing by 2% and O₃ in the Northern Hemisphere midlatitudes increasing by 2%. It is also interesting that the altitude of the maximum changes of OH and O₃ is higher in E50 than in A50. This is because the concentration of OH is higher in East Asia than North America in the troposphere (not shown), and when CH₄ emissions increase in East Asia, CH₄ could be fully oxidized and generate more O₃ in the troposphere. When CH₄ emissions increase in North America, the oxidation reaction of CH₄ and OH is not sufficient, so more CH₄ is transported into the stratosphere, where it is oxidized, and more O₃ is generated in the stratosphere.

O₃ is generated through the CH₄ oxidation in the presence of NO _x in the troposphere (e.g., Crutzen, 1973). During the 20th century, global background O₃ concentrations rose by at least a factor of two, due mainly to increases in CH₄ and NO _x emissions (e.g., Marenco et al., 1994; Wang and Jacob, 1998). Therefore, the distribution of CH₄ and NO _x in the troposphere plays an important role in generating O₃ (e.g., Fiore et al., 2008).

Figure 7 shows O₃ differences at different heights. There is an obvious local impact of CH₄ emission increases, and the maximum increase of O₃ is 5% in E50 and 4% in A50 at 500 hPa. In the upper troposphere at 200 hPa, in E50 the O₃ concentration increases by 6% in East Asia, higher than surrounding regions. In A50 the O₃ concentration increment in North America is higher than surrounding regions, but the O₃ does not increase significantly in the other lower latitude regions, and the O₃ increment in North America is only 4%, much smaller than E50. The O₃ increase is symmetrically distributed along the equator in G15.

Tropospheric O₃ is an important greenhouses gas, and as a chemically active gas it can react with numerous active substances. High concentrations of tropospheric O₃ are harmful to health, affecting the respiratory system. When CH₄ emissions increase by the same amount in East Asia and North America, the effect on tropospheric O₃ in the former is larger than the latter.

(Lang et al., 2012) found that the seasonal variation of O₃ change due to CH₄ emission increases in the troposphere is very weak, except near the tropical tropopause where O₃ has a maximum increase in July. As shown in Fig. 7, the local change of tropospheric O₃ is significant when CH₄ emissions increase in East Asia and North America; therefore, the O₃ changes in different months in East Asia and North America are interesting. The distribution of tropospheric O₃ concentrations in East Asia and North America are shown in Fig. 8a, and are in good agreement with (Wang et al., 2006). As shown in Fig. 8b, there is an obvious seasonal variation of O₃ change in East Asia when CH₄ emissions increase in East Asia in the upper troposphere (200 hPa). The maximum increment appears in August. The seasonal variation of O₃ change in North America in A50 shows that the maximum increment appears in April. In G15, the variation is small in different months and the increment of O₃ is large from May to September.

Figure 8.  (a) Annual average O₃ for different months simulated by the control experiment (E0) in East Asia (black) and North America (blue). Dotted and solid lines indicate simulations at 200 hPa (left vertical axis) and 500 hPa (right vertical axis). (b) Differences of annual average O₃ during different months of the sensitivity experiments compared with E0 at 200 hPa. (c) As in (b) but for 500 hPa. The results for E15 (black), E30 (blue), E50 (green), and G15 (red solid line) indicate the difference compared with E0 in East Asia. The results for A50 (orange) and G15 (red dotted line) indicate the difference compared with E0 in North America.

O₃ increases are more consistent at 500 hPa than at 200 hPa, as the maximum increment of O₃ concentration appears in August in East Asia and in July in North America. However, the seasonal variation of O₃ change in G15 is not obvious (Fig. 8c) in either East Asia or North America. This proves that seasonal variation of O₃ concentrations, shown in Fig. 8, are affected by CH₄ emission increases locally in the lower troposphere. Comparing Figs. 8a-c, the maximum O₃ increment appears when the original O₃ concentration is at its minimum in East Asia, so the percentage change of O₃ concentration is similar to Figs. 8b and c. Although the O₃ concentration increase is smaller in the middle troposphere than in the upper troposphere, the seasonal variation is also significant in the middle troposphere. The O₃ concentration is lower at 500 hPa than 200 hPa, so the percentage change of O₃ concentration is larger at 500 hPa than 200 hPa (not shown).

Since CH₄ oxidation by OH depends on temperature, when CH₄ emissions increase the largest O₃ concentration increment should be in summer. But in A50, the maximum increment appears in April at 200 hPa. The seasonal variation of O₃ change in A50 is inconsistent with the seasonal variation of O₃ concentration simulated by the control experiment shown in Figs. 8a and b. That is, the O₃ concentration change in A50 at 200 hPa is caused by transport processes. So, when CH₄ emissions increase in East Asia, its impact on the tropospheric O₃ can be noted throughout the troposphere. However, when CH₄ emissions increase in North America, the impact on the tropospheric O₃ can be noted only in the lower troposphere and the change of O₃ concentration in the upper troposphere is caused by transport processes.

A fully coupled chemistry-climate model (WACCM3) was used to investigate the effect of CH₄ emission increases, especially in East Asia and North America, on atmospheric temperature, circulation and O₃.

The results show that CH₄ emission increases can strengthen the westerly winds in the Northern Hemisphere midlatitudes, and accelerate the BD circulation, increasing the mass flux across the tropopause. However, the tropical BD circulation between 10^°S and 10^°N at 100 hPa becomes weaker when CH₄ emissions increase in East Asia, and stronger when CH₄ emissions increase in North America. When CH₄ emissions increase by 50% in East Asia and 15% globally, the stratospheric temperature cools by up to 0.15 K, and the stratospheric O₃ increases by 45 ppbv and 60 ppbv, respectively. When CH₄ emissions increase by 50% in North America the stratospheric O₃ increases by 60 ppbv, so it has a greater influence on stratospheric O₃ than the effect of a CH₄ emissions increase of the same amount in East Asia. CH₄ increases in East Asia and North America can reduce the tropospheric OH concentration (4% and 2%, respectively) and increase the tropospheric O₃ concentration (5% and 4%, respectively) in the Northern Hemisphere midlatitudes. When CH₄ increases in East Asia, the increase of the tropospheric O₃ concentration is largest in August. When CH₄ increases in North America, the increase of the O₃ concentration reaches a maximum in July in the middle troposphere, and in April in the upper troposphere. When CH₄ emissions increase in East Asia, the impact on tropospheric O₃ can be noted throughout the troposphere. However, when CH₄ emissions increase in North America, the impact on tropospheric O₃ is notable only in the lower troposphere and the change of O₃ concentration in the upper troposphere is caused by transport processes. The emissions scenarios used in this paper are provided by the IPCC at a resolution of 4^{°× 5°}, so the CH₄ emissions source is coarse. CH₄ reacts with CO, NO _x and other chemical substances, affecting tropospheric O₃ and stratospheric climate. Against the background of increasing pollutant emissions, the effect of a CH₄ emissions increase must be associated with CO, NO _x and other substances. Therefore, much more work is needed to fully understand the effect of CH₄ emission increases on both tropospheric and stratospheric climate in East Asia.