Kwon-Ho Lee

Estimation of potential source regions of PM2.5 in Beijing using backward trajectories

Possible source regions of the PM2.5 mass concentrations in Beijing, China have been identified using the Potential Source Contribution Function (PSCF) analysis based on the backward trajectory modeling. Five–day backward trajectories for the arrival of air masses to the ground observation station at Chaoyang Olympic Sports Center were calculated, using the Hybrid Single–Particle Lagrangian Integrated Trajectory (HYSPLIT) model in 2013. The PSCF results demonstrated that regional sources in different seasons could be one of the crucial contributors to PM2.5 mass concentrations in Beijing, especially in the winter season. Compared PSCF results with the MODIS fine–mode aerosol optical thickness (AOTfine), the regions showing high AOTfine especially the Southern Hebei, Northern Henan and Southwest Shandong are the significant potential PM2.5 contributors to Beijing. Environmental authorities may use the derived models and results for pinpointing the source areas, for improving air quality in Beijing City.

Retrieval of dust storm aerosols using an integrated Neural Network model

Dust storms are known to have adverse effects on public health. Atmospheric dust loading is also one of the major uncertainties in global climatic modeling as it is known to have a significant impact on the radiation budget and atmospheric stability. This study develops an integrated model for dust storm detection and retrieval based on the combination of geostationary satellite images and forward trajectory model. The proposed model consists of three components: (i) a Neural Network (NN) model for near real-time detection of dust storms; (ii) a NN model for dust Aerosol Optical Thickness (AOT) retrieval; and (iii) the Hybrid Single Particle Lagrangian Integrated Trajectory (HYSPLIT) model to analyze the transports of dust storms. These three components are combined using an event-driven active geo-processing workflow technique. The NN models were trained for the dust detection and validated using sunphotometer measurements from the AErosol RObotic NETwork (AERONET). The HYSPLIT model was applied in the regions with high probabilities of dust locations, and simulated the transport pathways of dust storms. This newly automated hybrid method can be used to give advance near real-time warning of dust storms, for both environmental authorities and public. The proposed methodology can be applied on early warning of adverse air quality conditions, and prediction of low visibility associated with dust storm events for port and airport authorities.

Investigation on a Haze Episode of Fine Particulate Matter using Semi-continuous Chemical Composition Data

In this study, semi-continuous measurements of mass, organic and elemental carbon (OC and EC), black carbon (BC), and ionic species concentrations were made for the period of April 03~13, 2012, at a South Area Supersite at Gwangju. Possible sources causing the high concentrations of major chemical species in observed during a haze episode were investigated. The measurement results, along with meteorological parameters, gaseous pollutants data, air mass back trajectory analyses and PSCF (potential source contribution function) results, were used to study the haze episode. Substantial enhancements of OC, EC, BC, , , , , and CO concentrations were closely associated with air masses coming from regions of forest fires in southeastern China, suggesting likely an impact of the forest fires. Also the PSCF maps for EC, OC, , and demonstrate further that the long-range transport of smoke plumes of forest fires detected over the southeastern China could be a possible source of haze phenomena observed at the site. Another possible source leading to haze formation was likely from photochemistry of precursor gases such as volatile organic compounds, , and , resulting in accumulation of secondary organic aerosol, and . Throughout the episode, local wind directions were between 200 and , where two industrial areas are situated, with moderate wind speeds of 3~5 m/s, resulting in highly elevated concentration of with a maximum of 15 ppb. The peak occurring in the afternoon hours coincided with maximum ambient temperature () and ozone concentration (~100 ppb), and were driven by photochemistry of . As a result, the pattern of variations in relation to wind direction, and concentrations, and the strong correlation between and (

A Retrieval of Vertically-Resolved Asian Dust Concentration from Quartz Channel Measurements of Raman Lidar

R^2

Classification of Dust/Non-dust Particle from the Asian Dust Plumes and Retrieval of Microphysical Properties using Raman Lidar System

The Light Detection and Ranging (Lidar) observation provides a specific knowledge of the temporal and vertical distribution and the optical properties of the aerosols. Unlike typical Mie scattering Lidars, which can measure backscattering and depolarization, the Raman Lidar can measure the quartz signal at the ultra violet (360 nm) and the visible (546 nm) wavelengths. In this work, we developed a method for estimating mineral quartz concentration immersed in Asian dust using Raman scattering of quartz (silicon dioxide, silica). During the Asian dust period of March 15, 16, and 21 in 2010, Raman lidar measurements detected the presence of quartz, and successfully showed the vertical profile of the dust concentrations. The satellite observations such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) confirmed spatial distribution of Asian dust. This approach will be useful for characterizing the quartz dominated in the atmospheric aerosols and the investigations of mineral dust. It will be especially applicable for distinguishing the dust and non-dust aerosols in studies on the mixing state of Asian aerosols. Additionally, the presented method combined with satellite observations is enable qualitative and quantitative monitoring for Asian dust.

Utilization of the depolarization ratio derived by AERONET Sun/sky radiometer data for type confirmation of a mixed aerosol plume over East Asia

The particle depolarization ratios were retrieved from the observation with a multi-wavelength Raman lidar at Gwangju, Korea (, ). The measurements were carried out on 24 February and 9 March 2004. Using the particle depolarization ratios, the non-dust aerosol particles were distinguished from the Asian dust plume, and the proportion of the non-dust particle to total dust plume was retrieved. The calculated proportion of the non-dust particle was used for the retrieval of backscatter coefficients at 355, 532, and 1064 nm and extinction coefficients at 355 and 532 nm of non-dust particles in the dust plume. Microphysical parameters of non-dust particles including single-scattering albedo at 532 nm were retrieved using retrieved optical values. The retrieved single-scattering albedo of non-dust particles was 0.92~0.95 below 1 km height and 0.82~0.91 above 1 km height on 24 February 2004 and on 9 March 2004.

Source contributions and potential source regions of size-resolved water-soluble organic carbon measured at an urban site over one year

ABSTRACT This article confirms the utilization of depolarization ratio derived by ground-based Aerosol Robotic Network (AERONET) Sun/sky radiometer data obtained during a high-PM10 episode at Gwangju, Korea (35.10° N, 126.53° E) in April 2009, in order to determine the nature and source of the atmospheric aerosol associated with this event. Integrated monitoring using satellite and depolarization light detection and ranging (lidar) data, together with model analysis, was also completed for the period of the high-PM10 event. The Sun/sky radiometer-derived particle depolarization ratio values are similar to the lidar-derived values, and these values highlight the effect of dust particles on aerosol observation. High particle depolarization ratios (12.5–14.2%) were observed when the aerosol plume transported from the west between 5 and 7 April. In contrast, lower particle depolarization ratios (5.8–9.8%) were detected when the aerosol plume was transported from the north on other observation days. Different optical properties are also shown according to the variation of depolarization ratio. High values in the real part of the refractive index (1.47–1.49 at 440 nm), lower values in the imaginary part of the refractive index (0.007–0.009 at 440 nm), and a high proportion of coarser particles were observed during the high depolarization ratio period. In contrast, the atmospheric aerosol transported from the north showed characteristics more commonly associated with smoke, with lower values in the real part of the refractive index (1.41–1.48 at 440 nm), higher values in the imaginary part of the refractive index (0.008–0.011), and a high proportion of fine particles. This indicates that the Sun/sky radiometer-derived depolarization ratio is a useful parameter when estimating the effect of dust particles during high-PM10 events.

Monitoring and visualizing the transport of atmospheric aerosols using satellite and ground based observations

In this study, 24 h size-segregated particulate matter (PM) samples were collected between September 2012 and August 2013 at an urban site in Korea to investigate seasonal mass size distributions of PM and its water-soluble components as well as to infer the possible sources of size-resolved water-soluble organic carbon (WSOC) using a positive matrix factorization (PMF) model. The potential source contribution function (PSCF) was also computed to identify the possible source regions of size-resolved WSOC. The seasonal average contribution of water-soluble organic matter to PM1.8 was in the range from 12.7 to 19.7%, but higher (21.0%) and lower contributions (8.9%) were observed during a severe haze event and an Asian dust event, respectively. The seasonal mass size distribution of WSOC had a dominant droplet mode peaking at 0.55 μm and a minor coarse mode peaking at 3.1 μm. The droplet mode WSOC was found to strongly correlate with oxalate, SO42-, NO3-, and K+, suggesting that in-cloud processes and biomass burning emissions are important sources of droplet mode WSOC. This finding was verified by the results obtained using PMF models. Secondary organic aerosols (oxalate + SO42- + NO3-) and biomass burning were the most important contributors (70.3%) to condensation mode WSOC. In the droplet mode, in-cloud processes and secondary NO3- (+biomass burning) were important sources of WSOC, contributing on average 46.4 and 25.9% to the WSOC, respectively. In the coarse mode, soil dust and secondary processes contributed 52.5 and 42.5% to the WSOC, respectively. The PMF analyses and PSCF maps of WSOC, SO42-, and K+ indicate that condensation mode WSOC was mostly influenced by the secondary organic aerosols and biomass burning from both local and long-range transported pollutants, while droplet mode WSOC was primarily the result of atmospheric processing during the long range transport of biogenic and anthropogenic pollutants from the eastern regions of China.

Synergetic analysis of springtime air pollution episodes over Gwangju, Korea

Two independent remote sensing, ground-based active (LIDAR) and passive (satellite and sun-photometry) techniques, offers a great potential to monitor atmospheric aerosols with its observational frequency. The synergetic use of these techniques with analytic differential method was devoted to provide multi-dimensional distribution and temporal variations for atmospheric aerosols. From satellites and ground based observations, the continuous measurements of the aerosol extinction coefficients and aerosol optical thickness (AOT) were conducted according to time and space over Northeast Asia (100°E-140°E, 20°N-50°N). In order to estimate the variation of aerosols, we present examples of the satellite retrieved AOT products with the LIDAR derived extinction profiles for a case study of polluted aerosol cases. The aerosol scale heights were also derived from the fitting of vertical extinction profile of LIDAR observations. These data were considered that the vertically resolved aerosols in a satellite observed pixels. Additionally, volumetric reconstruction of the aerosol distribution to visualize data. The exported data can allow near real-time aerosol monitoring and observation of the horizontal and vertical distributions in region of interest. Overall, the new product successfully retrieved and visualized the three-dimensional aerosol plume during its transport to the east Pacific from the Asian continent as the frequent seen in the both space and ground observations. Horizontal AOT data lead to a greater coverage of cloud-free aerosol loads in the study area, while the analytically resolved aerosol extinctions provide an information on aerosols vertical distributions. However, we note that the current version of retrieval algorithm can introduce significant uncertainties in the case of uplifted aerosols such as Asian dust storm observed in the study area.

Characteristics of water-soluble inorganic species in PM10 and PM2.5 at two coastal sites during spring in Korea

The characteristics of springtime aerosols, including their optical and microphysical properties, were analyzed for the months of March to May of 2009 in Gwangju (35.23°N, 126.84°E), Korea. A high Light Detection and Ranging (LIDAR)-derived aerosol depolarization ratio (δ) of 0.25±0.04 was determined on dust particles during the observation period. The Ångström exponent values of the 440-870nm wavelength pair (Å440-870) and single-scattering albedo at 675nm (Ω675) measured by a CIMEL sun/sky radiometer were 0.77±0.19 and 0.95±0.01, respectively. The elevated dust layers reached a maximum elevation of 4km above sea level. Anthropogenic/smoke particles that originated from highly populated/industrialized regions could be distinguished by their relatively smaller particle size (Å440-870 ranged between 1.33 and 1.36) and higher light-absorbing (Ω675 of 0.92±0.01) characteristics. These aerosols are mostly distributed at altitudes <1.2km. The root-mean-square deviation (RMSD) between the aerosol optical depth (AOD, τ) derived from LIDAR (τLIDAR) and from the CIMEL sun/sky radiometer (τCIMEL) varied with respect to the surface PM10 concentration. The RMSD between τLIDAR and τCIMEL was as low as 13% under lower PM10 concentration levels (<100μg/m3). In contrast, the RMSD between τLIDAR and τCIMEL increased three times (~31%) under high surface PM10 concentration levels (>100μg/m3). These results suggest that the accuracy of τLIDAR is influenced by specific atmospheric conditions, regardless of its uncertainty.