There has been significant progress in understanding aerosol composition and distributions in the troposphere during the last 20 years. This was in response to strong demands for (1) assessing the aerosol effects on the Earth radiative balance [e.g., Intergovernmental Panel on Climate Change, 2001] and (2) reducing air pollution associated with aerosol particles. One important key to this progress is the global coverage provided by satellite observations at high spatial resolution from instruments such as the Moderate-Resolution Imaging Spectroradiameter (MODIS) and the Multiangle Imaging SpectroRadiometer (MISR). These observations allow us to monitor global aerosol distributions and to diagnose aerosol column loadings in a retrospective fashion. In situ measurements, which allow direct measurements of chemical composition, are of high resolution but are sporadic and are suitable for more detailed process studies. An aerosol three-dimensional model can be used to link these two measurements by filling gaps in space and time. Several models have been developed with different representations of chemical composition, particle size distributions, chemical processes in both gas and aqueous phases, and interactions between aerosol particles and also between cloud drops (liquid and ice) and aerosol particles. These range from regional- and cloud-scale models [e.g., Binkowski and Roselle, 2003; Jacobson, 1997; Barth et al., 2001] to global models [e.g., Chin et al., 2003; Rasch et al., 1997].
 One of major constituents of aerosol is sulfate. Sulfate is produced predominantly via the oxidation of sulfur species (sulfur dioxide (SO2), dimethyl sulfide (DMS), hydrogen sulfide (H2S), and carbonyl sulfide (OCS)) in the atmosphere. Sulfur dioxide is a major precursor of sulfate in a tropospheric polluted environment. SO2 is oxidized to sulfate mainly by hydroxyl radical in the gas phase while it is oxidized mainly by hydrogen peroxide (H2O2), ozone (O3) or with iron (Fe(III)) through catalyzed reactions in the aqueous phase [Seinfeld and Pandis, 1998; Kreidenweis et al., 1997]. It has been found that the SO2 oxidation rate in the gas phase is much slower than that in the aqueous phase, i.e., 5% per hour in the gas phase and several hundred percent per hour (an equivalent gas-phase rate of SO2 conversion to sulfate) in the aqueous phase [Calvert et al., 1985]. Of the three major oxidation pathways in the aqueous phase, hydrogen peroxide is found to be the most effective oxidant under SO2 abundant conditions [Kreidenweis et al., 1997]. Clouds play a major role in the formation and redistribution of sulfate both locally and globally by providing a unique environment where liquid drops and ice crystals provide an effective medium for chemical reactions and cloud dynamical evolution provides a means for rapid vertical transport of chemical substances [e.g., Dickerson, 1987; Pickering et al., 1996].
 Sulfate in-cloud production not only adds sulfate mass to the local atmosphere but can also modify the aerosol size distribution. Observations show that populations of aerosol particles that have been processed in clouds have a distinct peak at 0. 1–1 μm diameter range in the size distribution [e.g., Hoppel and Frick, 1990]. The suggested mechanism for creating this peak in aerosol size distribution is the following. Cloud drops nucleate on cloud condensation nuclei (CCN). As the cloud evolves, the cloud drops absorb water-soluble chemical species, adding mass to the existing chemical species in the drops. When the cloud evaporates, the chemical species may be released as aerosol particles, which are larger in size than the original CCN. In case of SO2 and H2O2 scavenging by clouds, SO2 may be oxidized by H2O2 to produce sulfate. Consequently, aerosol particles processed in clouds have increased sulfate mass and produce an additional peak in the size distribution. Adding sulfate mass to CCN considerably alters the physicochemical and optical properties of CCN by enhancing the hygroscopy of aerosol particles. This leads to the modification of cloud properties and rain persistence, thereby affecting the local radiative and hydrological balance.
 East Asia is one of largest source regions of sulfur dioxide in the world, although sulfur dioxide emissions have decreased in recent years due to the slowdown of east Asian economic development [Streets et al., 2003]. Air masses influenced by these emissions may travel long distances due to the strong westerlies within the Japan jet stream during spring. There is evidence that the east Asian outflow modifies the air mass characteristics downwind and adversely affects the North Pacific and North American air quality [Jaeglé et al., 2003; Price et al., 2003]. As one of the NASA Global Tropospheric Experiment (GTE) efforts, Transport and Chemical Evolution Over the Pacific (TRACE-P) was conducted in spring 2001 to characterize the Asian chemical outflow and to determine its chemical evolution [Jacob et al., 2003]. Tu et al.  describe the SO2 distributions measured during the TRACE-P period under different meteorological conditions. SO2 concentrations appeared to be influenced (reduced) by clouds during the TRACE-P flights. Tu et al.  compared model results obtained from CFORS/STEM-2K1 [Carmichael et al., 2003] with the measured concentrations. The comparisons showed the model poorly represented the SO2 distributions for the cloudy cases whereas the modeled SO2 distributions were in good agreement with the measured values for clear sky cases. Crawford et al.  also found the observational evidence for cloud scavenging of SO2 and HNO3 during the TRACE-P period.
 The objective of this paper is to estimate the sulfate production via in-cloud processing in east Asia during the TRACE-P period using the University of Wisconsin Nonhydrostatic Modeling System (UWNMS). UWNMS forms the dynamical core of the regional component of Regional Air Quality Modeling System (RAQMS). RAQMS was used to estimate the tropospheric ozone budget over east Asia during the TRACE-P period [Pierce et al., 2003], also the carbon monoxide transport for the TRACE-P period [Kiley et al., 2003]. Hitchman et al.  also use the UWNMS to discuss the TRACE-P flight of 24 March with a focus on transport of alcohols and stratospheric ozone by the convectively induced circulation. For this study, the UWNMS is coupled with an aqueous sulfur chemistry model, which represents the oxidation of SO2 by H2O2 in the aqueous phase. Other competing processes in sulfate production such as gas phase reactions and the oxidation of other sulfur species are considered to be of secondary importance for sulfate in-cloud production and not included in the model. The current version of the sulfur model is part of a broader development effort to incorporate full aerosol chemistry into RAQMS, which currently has full tropospheric gaseous chemistry components. Two global models provide initial and lateral boundary conditions for the regional model simulations. The global component of RAQMS [Pierce et al., 2003] provides H2O2 distributions for the initial and boundary conditions of the regional model. Because RAQMS does not employ sulfur chemistry, the Global Ozone Chemistry Aerosol Radiation and Transport (GOCART) model [Chin et al., 2000a, 2000b, 2002, 2003] is used for SO2 and sulfate initial and boundary conditions of the regional model.
 In sections 2 and 3the components of models used are briefly described. In section 4.1 we characterize the regional model performance of predicting the atmospheric components involved in the sulfate in-cloud production, such as the distributions of SO2, sulfate, H2O2, liquid water in clouds, and precipitation by comparing these components with TRACE-P aircraft measurements and satellite observations. We then describe a case study for 24 March 2001, when the air influenced by a deep convective storm was advected from the east coast of China to the south of the Japan islands (section 4.2.1). We conclude by estimating the contribution of the sulfate in-cloud production to the sulfate budget for the TRACE-P period (sections 4.2.2 and 5).