Abstract. To improve our knowledge of urban meteorology, including those processes applicable to high-resolution meteorological models in the Seoul Metropolitan Area (SMA), the Weather Information Service Engine (WISE) Urban Meteorological Observation System (UMS-Seoul) has been designed and installed. The UMS-Seoul incorporates 14 surface energy balance (EB) systems, 7 surface-based threedimensional (3-D) meteorological observation systems and applied meteorological (AP) observation systems, and the existing surface-based meteorological observation network. The EB system consists of a radiation balance system, sonic anemometers, infrared CO 2 /H 2 O gas analyzers, and many sensors measuring the wind speed and direction, temperature and humidity, precipitation, and air pressure. The EBproduced radiation, meteorological, and turbulence data will be used to quantify the surface EB according to land use and to improve the boundary-layer and surface processes in meteorological models. The 3-D system, composed of a wind lidar, microwave radiometer, aerosol lidar, or ceilometer, produces the cloud height, vertical profiles of backscatter by aerosols, wind speed and direction, temperature, humidity, and liquid water content. It will be used for highresolution reanalysis data based on observations and for the improvement of the boundary-layer, radiation, and microphysics processes in meteorological models. The AP system includes road weather information, mosquito activity, water quality, and agrometeorological observation instruments. The standardized metadata for networks and stations are documented and renewed periodically to provide a detailed observation environment. The UMS-Seoul data are designed to support real-time acquisition and display and automatically quality check within 10 min from observation. After the quality check, data can be distributed to relevant potential users such as researchers and policy makers. Finally, two case studies demonstrate that the observed data have a great potential to help to understand the boundary-layer structures more deeply, improve the performance of high-resolution meteorological models, and provide useful information customized based on the user demands in the SMA.

No abstract

No abstract

Severe PM 2.5 air pollution over the Asian continent is occasionally transported across the East China Sea by the westerly winds to Japan, continuing for long distances over the Pacific Ocean. Despite such polluted air masses causing health issues, conventional models tend to underestimate levels of organic aerosols (OA) and PM 2.5 . Here, PM 2.5 and its major components recorded during three field campaigns carried out at Fukue Island (32.75°N, 128.68°E), Japan (spring 2009), Rudong (32.25°N, 121.37°E), China (spring 2010), and Jeju (33.35°N, 126.39°E), Korea (autumn 2012) around the East China Sea were used to test the performance of the Weather Research and Forecasting-Chem/ATRAS-MOSAIC model. Overall, model performance was improved by introducing chemical aging represented by a volatility basis-set scheme, whereby median values of the model/observation ratio for OA were raised to 0.34-1.28 from 0.30-0.35 in the case of conventional settings. In particular, the levels of OA at the Fukue site and daytime buildup of the OA levels at all three sites were reproduced by the model. OA levels were still sometimes underestimated. This suggests that either emission rates of organic precursors are being underestimated or other pathways of OA formation are also important. Our analysis also indicates that this region is characterized by high OH concentrations, promoting chemical aging. The predictions of PM 2.5 levels in the model also improved, with median values of the model/observation ratio shifting from 0.67-0.91 to 0.68-0.95, when chemical aging of OA was taken into account.

Abstract. Despite recent progress of numerical air quality models, accurate prediction of fine particulate matter (PM2.5) is still challenging because of uncertainties in physical and chemical parameterizations, meteorological data, and emission inventory databases. Recent advances in artificial neural networks can be used to overcome limitations in numerical air quality models. In this study, a deep neural network (DNN) model was developed for a 3 d forecasting of 6 h average PM2.5 concentrations: the day of prediction (D+0), 1 d after prediction (D+1), and 2 d after prediction (D+2). The DNN model was evaluated against the currently operational Community Multiscale Air Quality (CMAQ) modeling system in South Korea. Our study demonstrated that the DNN model outperformed the CMAQ modeling results. The DNN model provided better forecasting skills by reducing the root-mean-squared error (RMSE) by 4.1, 2.2, and 3.0 µg m−3 for the 3 consecutive days, respectively, compared with the CMAQ. Also, the false-alarm rate (FAR) decreased by 16.9 %p (D+0), 7.5 %p (D+1), and 7.6 %p (D+2), indicating that the DNN model substantially mitigated the overprediction of the CMAQ in high PM2.5 concentrations. These results showed that the DNN model outperformed the CMAQ model when it was simultaneously trained by using the observation and forecasting data from the numerical air quality models. Notably, the forecasting data provided more benefits to the DNN modeling results as the forecasting days increased. Our results suggest that our data-driven machine learning approach can be a useful tool for air quality forecasting when it is implemented with air quality models together by reducing model-oriented systematic biases.