Asian chemical outflow to the Pacific in spring: Origins, pathways, and budgets

,


Rapid industrialization of eastern Asia is expected to have important implications for global atmospheric chemistry over the next decades [Berntsen et al., 1996] and also perhaps for surface air pollution over North
1997], further complicates the composition of the outflow and its evolution [Zhang and Carmichael, 1999]. [Jacob et al., 1999;Berntsen et al., 1999;Iqenger et al., 2000]. More interest needs to be given to the Asian outflow in terms of quantification of the flux exported and the mechanisms involved in the export. Using a regional model, Carmichael et al. [1998] found that stratospheric intrusions as well as Asian outflow contribute significantly to ozone concentrations in surface air over Japan. ¾ienger et al. [2000] showed that convergence over Asia makes a major contribution to the Asian outflow over the western Pacific. In the present study, we focus on quantifying the Asian export of CO, NO• and ozone, which are of particular interest for driving global atmospheric chemistry. Our simulation uses the GEOS-CHEM global 3-D model of tropospheric chemistry with assimilated meteorological observations for 1994, and our evaluation of model results focuses on PEM--West B observations. We identify the major pathways and meteorological drivers for export of Asian pollution, esti-mate the role of chemical processing over the continent, and quantify the contributions from various sources including biomass burning and intercontinental transport to the Asian outflow over the western Pacific.

Model Description
The  ]. Lightning emission of NOx occurs in conjunction with deep convective events in the GEOS data following the parameterizations of Price and Rind [1992] and Pickering et al. [1998]. Advection is computed with a flux-form semi-Lagrangian method described by Lin and Rood [1996]. Moist convection is computed using the GEOS convective, entrainment, and detrainment mass fluxes as described by Allen et al. [1996a, 1996b] to January 1994 effectively removes the effect of initial conditions, and results from February-March 1994 are used to compare with the observations. We also present in this paper a CO simulation with the original 2øx2.5 ø horizontal resolution to better resolve the structure in the Asian outflow and to diagnose any failures of the coarser resolution in representing major features of transport. This CO-only simulation uses eight tagged CO tracers to resolve source regions contributing to Asian outflow: four tracers for fuel combustion including fossil fuel and wood (North America, Europe, Asia, and rest of the world) and four tracers for biomass burning (South America, Africa, Asia, and rest of the world). The corresponding domains are given in Figure 2. Loss of CO by reaction with OH and production of CO from oxidation of isoprene and methane by OH (using molar yields of 1.25 and 1, respectively) are calculated using OH monthly mean fields generated with the standard simulation described previously. The source of CO due to oxidation of hydrocarbons other than methane and isoprene is relatively small and is ignored. By summing the concentrations of all CO tracers we reproduce closely the CO concentrations obtained in the standard full-chemistry simulation. 180 ø decrease, while incursions of warmer and tropical air from the south become more frequent. The warming leads to more frequent development of convective thunderstorms, especially in Southeast Asia [Nieuwolt, 1977]. Figure 5 shows the presence of a convergence zone over central China where air masses from the north, driven by monsoon winds, encounter oceanic air masses coming from the south. As expected, the convergence zone becomes more apparent as the winter monsoon weakens and the regime of summer monsoon slowly starts to establish. As will be discussed later, this convergence zone plays an important role in the springtime export of pollution from the Asian continent. In 2) from anthropogenic activities and biomass burning. The main export pathway for Asian pollution is to the Pacific in the westerly flow north of 25øN. Figure 5 shows northerly surface winds along the coast of China that reach southern latitudes but the winds shift to westerly above 1 km altitude.

Maps of horizontal CO fluxes at individual levels indicate
that little mass is carded in the northerly surface flow, as is apparent from Figure 7b, and part of this flow is eventually recirculated over the continent by anticyclonic circulation over Southeast Asia. Figure 7 shows strong southwesterly CO fluxes over China from 20øN to 30øN which result from the colocation of high emissions with the convergence zone described previously (Section 4.1). This convergence results in an upward flux of CO (see Figure 8) which lifts the pollution above the boundary layer into the free troposphere where it is caught by the strong westerlies. We thus find that the strongest export flux of Asian CO to the western Pacific is at 4 km altitude (Figure 8b) even though the highest concentrations along the Pacific rim are found below 2 km altitude (Figure 4a). We investigated separately the contributions of anthropogenic and biomass burning sources to the export of Asian CO to the western Pacific. Biomass bur 'ng CO, mainly emitted in Southeast Asia, is transported toward the conver-gence zone where it is uplifted into the free troposphere and then carded by the westefiies. Little biomass burning CO is exported over the ocean in the boundary layer (Figure 8d). In contrast, CO from fossil fuel combustion, which is emitted at more northerly latitudes, shows substantial export in the boundary layer by the monsoon winds, especially at latitudes higher than 35øN. Even for fossil fuel CO, however, most of the export is in the lower free troposphere (Figure 8c). In our model, during February-March, deep convective events are largely restricted to Southeast Asia, and thus mainly contribute to export of biomass burning CO. Large-scale convergence rather than convection is the principal driver for ventilation of Asian pollution from the boundary layer to the free troposphere in our model during February-March. (22øN, 122øE). Their measurements are shown in Figure 9, together with GEOS meteorological data for temperature and pressure at the site and corresponding model results for CO.

As discussed in section 4.1, the synoptic weather pattern in Asia in late winter-early spring is dominated by the passage of cold fronts, each cold front being followed by strong outbreaks of cold air masses. At the passage of each front, Liu et al. [1997] observed a shift of wind direction from
north-northeasterly to south-southwesterly, bringing marine tropical air with low CO to the site. After the passage of the front, they observed a sharp increase of CO due to transport from the Asian continent. Frontal passages in the GEOS meteorological data, as diagnosed from pressure drops followed by temperature drops (Figure 9a), match the dates identified by Liu et al. [1997] i.e., February 6-7, 11-12, 14-15, i 9-20, 23-24, and March 7-8, 12-13, 19-20, 22-23. The model reproduces well most of the events of high CO concentrations observed at the site, as shown in Figure   10 major events can be identified in our model simulation (Figure 9c). Our analysis is consistent with that of Yienger et al. [2000], who proposed that an important mechanism for the export of pollution from Asia is the development of low-pressure baroclinic systems over Asia.

Transport to Asian Outflow
The chemical outflow from the Asian continent to the western Pacific includes contributions from other continents besides Asia. Figure 11 shows the total column concentrations of each tagged CO tracer for February-March 1994, and Table 1 summarizes their contributions to the total CO burden in the Northern Hemisphere and to the Asian outflow (defined as the flux through a wall located at 140øE between 20 ø and 50øN). We discuss here the contributions of different geopolitical source regions as those originating from direct CO emissions only.

Export of NOv and Ozone From Asia
We isolate the contribution of Asian emissions to the budgets of total reactive nitrogen oxides (NOv) and ozone in the Asian boundary layer for the PEM-West B period by subtracting the background terms (given by a simulation with no Asian emissions) from the terms obtained in our standard simulation. We refer to the resulting budgets as those of "Asian" NO v and ozone. Although this method is only approximate because tropospheric chemistry is not linear, it allows us to isolate the fate of NO x and other compounds emitted over Asia. The budget of ozone is actually computed for the extended odd oxygen family Ox = 03 + NO2 +2xNO3 + 3xN205 + HNO4 + HNO3 + peroxyacylnitrates to account for rapid chemical cycling within the species in that family. Considering that 03 accounts for more than 95% of O•, the budget of 03 and O• can be considered equivalent.     Table 3 shows the budget of ozone in the boundary layer over Asia due solely to Asian emissions of precursors (hydrocarbons, CO, NO•:). Background Oa is subtracted following the procedure previously described. Fossil fuel

Summary and Conclusion
We used a global 3-D model of tropospheric chemistry driven by assimilated meteorology (GEOS-CHEM model) Mean The contribution of intercontinental transport of pollution to the Asian outflow over the western Pacific was examined in the model by tagging CO emitted from different source regions. We find that both anthropogenic sources in Europe and biomass burning in Africa make major contributions to the Asian outflow, with distinct geographical signatures.
European pollution dominates the Asian outflow in the lower troposphere at high latitudes, while African pollution is important in the upper troposphere at all latitudes. A budget analysis for the fate of NO x emissions in East Asia during February-March 1994 indicates that 5% is exported out of the Asian boundary layer as NOx and 8% as PAN. In comparison, it has been estimated previously that 20% of NO x emitted in the United States in spring is exported as NOx or PAN. The lower export efficiency for Asian emissions reflects higher aerosol concentrations that promote heterogeneous conversion of NOx to HNO3 by hydrolysis of N20,5. We find that production of 03 over East Asia and its export to the global atmosphere are much higher than for the United States because of the lower latitude of the Asian sources.
The observations from the PEM-West B mission have been of considerable value as an initial test of our simulation of Asian outflow to the Pacific. This mission was exploratory, however, and it does not provide the data necessary for identifying the source regions contributing to the outflow or for establishing the outflow mechanisms. Further work is needed to test several of the hypotheses presented in the present paper regarding the springtime Asian outflow, notably that (1) lifting of pollution ahead of cold fronts is a major mechanism for export of pollution from China and Southeast Asia to the western Pacific; (2) biomass burning and fossil fuel combustion make contributions of comparable magnitude to the export of Asian CO, NOy species,