Chengcai Li

Global monitoring of air pollution over land from the Earth Observing System‐Terra Moderate Resolution Imaging Spectroradiometer (MODIS)

[1] Moderate Resolution Imaging Spectroradiometer (MODIS) measurements (7 channels: 0.47-2.1 μm, 250-500 m resolutions) provide us with new insights into the characteristics of global aerosols. MODIS retrieves not only aerosol loading but also the fraction of fine mode particle. In this paper we demonstrate MODIS capability for use in monitoring global, regional, and local air pollution. Three case studies in northern Italy, Los Angeles, and Beijing showed the conclusive results of applying MODIS-derived aerosol optical depths (T a ) to regional and local air pollution in terms of accuracy (ΔT a = ±0.05 ± 0.2τ a ) and spatial sensitivity of the retrievals. Under stagnant condition, accumulated aerosol abundance can reach T a > 1 (at 0.55 μm) before being removed by wind or precipitation. The correlation found between Aerosol Robotic Network (AERONET) daily averaged T a and 24-hour PM 10 (particulate matter with diameter <10 μm) concentration (μg/m 3 ) in northern Italy is encouraging with correlation coefficient ∼0.82. The derivation of PM concentration from satellite measurements may be possible once we know the detailed aerosol vertical distribution. To compare aerosol loading in different regions of the globe, we choose the two most populated regions (eastern China and India) and the two most industrialized regions (the eastern United States/Canada and western Europe). The time series of MODIS monthly mean T a from July 2000 to May 2001 depicts a strong seasonal variation with maxima in the spring/ summer and minima in the winter. The clear separation between (1) the eastern United States/Canada and western Europe and (2) eastern China and India shows that the T a values in (2) are 50% to 2-3 times higher compared to those in (1). The enhancements of aerosol loading were due to smoke as originated from Montana/Idaho forest fires transported to the eastern United States in late August 2000 and dust outbreaks from Taklimakan and Gobi Deserts to eastern China as well as smoke from Southeast Asia to southern China in February-April 2001.

Retrieval, validation, and application of the 1-km aerosol optical depth from MODIS measurements over Hong Kong

The Moderate Resolution Imaging Spectroradiometer (MODIS) aerosol retrieval algorithm was developed to derive aerosol properties at a global scale, suitable for climate studies. Under favorable conditions (clear sky and over dark surfaces), the standard 10/spl times/10 km MODIS aerosol products are also useful on regional scales to monitor aerosol distributions and transports. However, the 10-km resolution is insufficient to depict aerosol variation on local or urban scales, due to inherent aerosol variability as well as complex surface terrain. In this study, we have modified the MODIS algorithm to retrieve aerosol optical depth (AOD) at 1-km resolution over Hong Kong, a city of just over 1000 km/sup 2/ with very complex surface features. Accompanied by the increased spatial resolution are new aerosol models derived with single-scattering albedo (SSA) around 0.91-0.94 to accommodate higher aerosol absorption encountered in Hong Kong than that was presumed for MODIS standard products (SSA/spl sim/0.97) over the region. The derived AOD data are compared to handheld Microtops II sunphotometer observations at the Hong Kong University of Science and Technology and other locations across Hong Kong. Retrieval errors within 15% to 20% of sunphotometer measurements are found. Moreover, when compared with the standard 10-km AOD products, the 1-km AOD data are much better correlated with PM/sub 10/ measurements across Hong Kong, suggesting that the new 1-km AOD data can be used to better characterize the particulate matter distribution for cities like Hong Kong than the MODIS standard products.

Assessing Long-Term Trend of Particulate Matter Pollution in the Pearl River Delta Region Using Satellite Remote Sensing

Serious particulate matter (PM) pollution problems in many polluted regions of China have been frequently reported in recent years. Long-term exposure to ambient PM pollution is significantly associated with adverse health effects. Characterizing the long-term trends and variation in PM pollution is a basic requirement for evaluating long-term exposure and for guiding future policies to reduce the effects of air pollution on health. However, long-term, ground-based PM measurements are only available at a few fixed stations. In this study, an algorithm is developed and validated to estimate PM concentrations based on the satellite atmospheric optical depth with 1 km spatial resolution. The long-term trends of PM10 concentrations in the entire Pearl River Delta (PRD) region and different cities are quantified and discussed. From 2001 to 2013, the PM10 pollution of the entire PRD region was dominated by a decreasing trend of -0.15 ± 0.23 μg/m(3)·yr. This decreasing PM10 trend was apparent over 75% of the PRD area, with the most significant decreases observed in the center of the region. However, the remaining 25%, mostly located in the outskirts of the region, showed an increasing PM10 trend. This overall decreasing trend indicates the effectiveness of the control measures applied in the past decade for the primary pollutants.

Changes in surface aerosol extinction trends over China during 1980–2013 inferred from quality‐controlled visibility data

Pollution in China has been attracting extensive attention both globally and regionally, especially due to the perceptually worsening “smog” condition in recent years. We use routine visibility measurements from 1980 to 2013 at 272 World Meteorological Organization stations in China to assess the temporal changes in the magnitude and the sign of pollution trends. A strict and comprehensive quality control procedure is enforced by considering several issues not typically addressed in previous studies. Two methods are used to independently estimate the trend and its significance level. Results show that, in general, a strong increase in aerosol extinction coefficient over the majority of China is observed in the 1980s, followed by a moderate decrease in the 1990s, another increase in the 2000s, and a shift to decrease since around 2006 for some regions. Seasonally, winter and fall trends appear to be the strongest, while summer has the lowest trend.

Remote sensing precipitable water with GPS

The principle and method of remote sensing precipitable water (PW) by the Global Positioning System (GPS) are introduced. A series of tropospheric delay have been obtained from the dual frequency receiver’s data of global tracking stations in East Asia and precise satelllte ephemeris of IGS from July 31 to August 20, 1997. The continuous PW at the interval of 30 min have been estimated at Shanghai and Wuhan GPS sites. The results are compared with PW from conventional radiosonde during the same period, and the RMS is about 0.50 cm.

An intercomparison of long‐term planetary boundary layer heights retrieved from CALIPSO, ground‐based lidar and radiosonde measurements over Hong Kong

The planetary boundary layer height (PBLH) is a very important parameter in the atmosphere, because it determines the range where the most effective dispersion processes take place, and serves as a constraint on the vertical transport of heat, moisture and pollutants. As the only space-borne lidar, Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) onboard Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) measures the vertical distribution of aerosol signals and thus offers the potential to retrieve large-scale PBLH climatology. In this study, we explore different techniques for retrieving PBLH from CALIPSO measurements, and validate the results against those obtained from ground-based micropulse lidar (MPL) and radiosonde (RS) data over Hong Kong, where long-term MPL and RS measurements are available. Two methods, namely maximum standard deviation (MSD) and wavelet covariance transform (WCT) are used to retrieve PBLH from CALIPSO. Results show that the RS- and MPL-derived PBLHs share similar interannual variation and seasonality, and can complement each other. Both MSD and WCT perform reasonably well compared with MPL/RS products, especially under sufficient aerosol loading. Uncertainties increase when aerosol loading is low and the CALIPSO signal consequently becomes noisier. Overall, CALIPSO captures the general PBLH seasonal variability over Hong Kong, despite a high bias in spring a low bias in summer. The spring high bias is likely associated with elevated aerosol layers due to transport, while the summer low bias can be attributed to higher noise level associated with weaker aerosol signal.

Validation of MODIS AOD products with 1-km resolution and their application in the study of urban air pollution in Hong kong

Remote sensing products of aerosol from MODIS have been released by NASA for a long time, and the Level 2 products showed to be very useful in monitoring regional aerosol pollution pattern and tracing pollutant transports. However, for a city, with an area of several hundreds square-kilometers and complicated terrain, the products with 10 km resolution are not enough especially in depicting the detail particulate matter (PM) distribution in the urban area. Recently, the aerosol optical depth (AOD) retrieval method proposed by NASA has been used to obtain the 1km resolution products in Hong Kong. Aerosol model was improved by sun-photometer observations and the calculation of a radiation transfer model, and finally looking up tables were created for real time aerosol products retrieval. The AOD products were validated by the sun-photometer observations at the HKUST. It was found the relative bias between the satellite products and the ground observations was within the range of about 20%, which was mostly equal to the estimation of NASA for their 10km level 2 products in most AERONET sites. The good results are mainly because of the most regions of Hong Kong are covered by dense dark vegetations (DDV) with very low surface reflectance in visual and near-infrared satellite channels. This high-resolution product was used in the study of air pollution in Hong Kong, and it was found the 1km products were more useful to describe the local urban PM pollutant distribution than the 10 km Level 2 products.

A new way of using MODIS data to study air pollution over Hong Kong and the Pearl River Delta

Aerosols are the main air pollutant in Asia. In this paper, the MODIS level 2 aerosol optical depth (AOD) products derived by NASA were validated with in situ sun-photometer observations over Hong Kong (HK). The MODIS AOD values were correlated with mass concentrations of respirable suspended particulates (RSP) measured at air quality monitoring stations over HK and Macau. Correlation between RSP and AOD were found to be statistically significant, suggesting that the satellite data is very useful for aerosol-related air pollution studies. Compared with concentrations measured from ground-based air quality monitoring networks, the AOD data cover a much larger area and have much better spatial resolution. Combining with meteorological information, the AOD data also proved to be very useful for the understanding of RSP variations at air quality monitoring stations. An example of using AOD data to help understand a pollution event over the PRD will be presented. Finally, monthly-mean distributions of AOD over Eastern China showed a distinct local maximum over the PRD, separated from high AOD areas to the north, suggesting that the aerosol problem over the PRD are mostly regional. Remote-sensing from space has provided a new and powerful way to study air pollution. To fully utilize this technique for air quality studies, the combination of a lidar and an X-band satellite receiver (for the MODIS data) is recommended. The AOD fields are vertically integrated products, together with the vertical profiles of extinction coefficients provided by a lidar, the surface distribution of aerosol could be derived.

Long-term characteristics of satellite-based PM 2.5 over East China

With the explosive economic development of China over the past few decades, air pollution has become a serious environmental problem and has attracted increasing global concern. Using satellite-based PM2.5 data from 2000 to 2015, we found that the temporal-spatial variation of PM2.5 in East China is characterized by high concentrations in the northern part and low concentrations in the southern part of East China, and by being seasonally high in autumn and winter but low in spring and summer. We also found that the regional average PM2.5 concentration shows an approximative peak pattern over the last 16years, with the highest, 60.13μgm-3, and the lowest, 46.18μgm-3, occurring in 2007 and 2000, respectively. Despite obviously diminishing heavy polluted regions with a PM2.5 of >80μgm-3 after 2011, those cells dominated by natural background have still not recovered back to the clean level of 2000. These characteristics are valuable information to analyze the relative contributions of anthropogenic emissions and atmospheric conditions to the temporal-spatial variation characteristics of PM2.5.

The Evolution of Springtime Water Vapor Over Beijing Observed by a High Dynamic Raman Lidar System: Case Studies

Raman lidar is an effective technique to retrieve the vertical distribution of atmospheric water vapor. For the first time, we present water vapor profiles retrieved by a high dynamic Raman lidar system over the Beijing area for representative cases in spring 2014, within the framework of the Aerosol Multi-wavelength Polarization Lidar Experiment project. In springtime, water vapor content over Beijing is generally low but with a strong daily variability. Its evolution is strongly coupled with winds and aerosols, with clouds also exerting a distinct impact. Northwesterly winds is found to be the most important factor impacting the temporal variability of water vapor mixing ratio (WVMR), and WVMR is found to be negatively correlated with wind speed. Moreover, we find that clouds tend to cause significant increases in the standard deviation of WMVR measurement, and relative humidity sharply increase below the cloud base. During a typical pollution episode, water vapor strongly covaries with aerosols due to hygroscopic growth effect and transport mechanism. Both water vapor and aerosols exhibit the highest variability within the planetary boundary layer (PBL), where the development and dissipation of haze mainly occur. Within the PBL, water vapor and aerosol concentration demonstrate different evolution features at different altitudes during the haze process, with a delayed increase and early decrease for higher altitudes. Back trajectory analysis using the hybrid single-particle Lagrangian trajectory model indicates that this phenomenon is most likely associated with different sources of the air mass at different altitudes.