Atmospheric Chemistry & Modeling

Global Atmospheric Environment and Human Impacts

• Our ACM research group takes an interdisciplinary approach to studying how human activities lead to globalizing air pollution and its consequences on air quality, climate and the ecosystems.

• We study emissions, transport and transformation of air pollutants, with a particular focus on China's pollution and its interactions with other regions.

• We use a suite of research tools from various disciplines, including economic-emission statistics, satellite remote sensing and atmospheric modeling.

• Our research topics include consumption-based pollution accounting, satellite remote sensing, and chemical transport model development.


Globalizing air pollution - Studying the influences of atmospheric transport and economic trade on air pollution, health, climate, and the ecosystems

Satellite remote sensing - Developing advanced high-resolution satellite remote sensing products to study air pollution and emissions

Modeling - Developing advanced high-resolution model to study physical, chemical and transport processes of air pollutants, including multi-scale interactions

Pollution-climate interactions - Combining model simulations and measurements to understand interactions between air pollution and climate

Energy - Assessing the potentials, limitations and environmental impacts of fossil fuels and renewable energy (wind, solar) for China and the world


At Harvard, I was working with Prof. Michael McElroy on Chinese environmental issues under the Harvard China Project. Below is a summary of the work I have been involved since I started to work at Harvard. Some of projects are continuing.

Improving the PBL representations in GEOS-Chem: Improving the CTM according to the best understanding of atmospheric physical, chemical and transport processes is essential to all model applications. The current standard version of GEOS-Chem assumes a full-mixing PBL, where concentrations, emissions and dry depositions of pollutants are evenly distributed within the PBL at any time. This assumption, however, overestimates the vertical mixing when the PBL is stable (e.g., at night) or neutral. To improve the PBL mixing representation in the model, I have implemented an alternate 'non-local' mixing scheme (Holtslag and Boville, 1993) that is capable of simulating the vertical mixing according to different states of PBL. The non-local PBL scheme has been shown to significantly improve the simulation on the diurnal variation of surface ozone concentrations. (paper) It also has significant impacts on top-down constraint of surface NOx emissions using satellite retrievals of tropospheric NO2 column. (paper)

Developing a new top-down approach to constrain NOx emissions: Current bottom-up estimations of surface pollutant emissions over China contain large uncertainties due to the lack of information on emission activity and emission factor. Therefore it is important to improve the understanding of surface emissions using other sources of information. One particularly useful approach is to use the CTM to constrain the emission sources according to the observed concentrations of pollutants (from satellite, aircraft, and ground stations) -- this is the so-called 'top-down' estimation. One of my current work is to constrain surface NOx emissions, including seasonal variation, using satellite remote sensing of NO2. I have developed a new approach by incorporating multiple satellite instruments observing the Earth at different times of day and explicitly accounting for diurnal variations of NOx emissions and concentrations. (paper)

A critical source of uncertainty in the top-down estimation is the inaccuracies in observations (e.g., from satellite instruments) and model representations of atmospheric processes. (An example is the full-mixing assumption in the standard GEOS-Chem.) This is because the top-down approach assumes no systematic errors in observations and models, and attributes all discrepancies between the observed and the modeled pollutant concentrations to inaccuracies in surface emissions. Also, the top-down estimation on one particular pollutant may be biased if emissions of other pollutants are not constrained simultaneously, due to the non-linear interactions of different pollutants in the atmosphere. Hence, one important aspect of my top-down estimation is to analyze the uncertainties due to such limitations in the top-down approach. (paper)

Evaluating the effectiveness of emission controls on particulate air pollution: Controls on anthropogenic sources of air pollutants are the key to improve the atmospheric environment. The Chinese government has moved aggressively since 2005 to reduce emissions of a number of pollutants including PM and SO2. The effect on atmospheric aerosol loading is unclear, particularly because of secondary sources from emissions of NOx, NMVOC and NH3 that lack stringent regulations. Satellite measurements of aerosol optical depth, together with surface measurements of mass concentrations of PM10, provide useful information to evaluate the effectiveness of emission controls on aerosols at different size ranges. Based on these information, it is found that recent emission controls have not been successful in reducing concentrations of aerosols at smaller size range in more industrialized regions of China. The lack of success is attributed to the increasing importance of secondary aerosols formed from NOx, NMVOC and NH3. (paper)

Analyzing air pollution changes due to natural experiments: In the policy-making process, a key factor for effective environmental policy is the adequate understanding of the relationship between pollutant emissions and concentrations. The emission-concentration relationship, however, is often non-linear and dependent on factors such as regions, meteorology, and relative abundances of different pollutants/precursors. Some natural experiments, such as the short-term emission regulation during the 2008 Beijing Olympics and Paralympics, provide valuable information for analyzing the emission-concentration relationship and evaluating the policy effectiveness for corresponding regions. In this case, one challenge is to separate the effects of emission reduction from the effects of meteorology, which is suggested to have major impacts on air quality during the time period. Therefore another of my current work is to use the CTM to analyze the observed changes in pollutant concentrations (particularly for NO2) during such natural experiments and delineate the effects of emission regulation.

Analyzing the potential for electricity generated from wind and solar radiation: Renewable energy is a future for sustainable economy and society. The intensive use of fossil fuel for energy has led to significant environmental problems including air pollution and climate change, which may be solved by switching to renewable energy sources. However, potentials of renewable energy are still a subject of on-going research. Currently, I am working on analyzing the potentials of electricity generated from wind and solar radiation for China, using wind and solar radiation data from 1982 to 2006 at high spatial (30km) and temporal (3-hour) resolutions derived from regional climate models.

Other interests: I am also interested in tropospheric aerosol transport and potential future changes in response to climate and emissions changes, particularly in the transport and transformation of Asian, especially Chinese, dust and anthropogenic aerosols and resulting impacts on downwind regions, including North America.


From August 2003 to April 2008, I worked with Prof. Donald Wuebbles in the Department of Atmospheric Sciences, University of Illinois at Urbana-Champaign on research of surface air quality at regional and urban scales, primarily ozone and other gaseous pollutants over the United States and China. I was involved in two key projects assessing potential impacts of global and regional climate and emissions on U.S. surface air quality initiated by the U.S. EPA. I also worked with Prof. Wuebbles, ISWS, and other research groups on two recent climate change impact assessments for the Northeast U.S. and for the Chicago region.

Using a state-of-the-art global chemical-transport model (CTM), Model for OZone And Related chemical Tracers (MOZART-2.4), I investigated a variety of important science issues related to surface ozone distributions and changes, including ozone diurnal variations, background continental ozone levels and short- and long-range transport, and projections and uncertainties of future ozone changes driven by climate changes and precursor emissions perturbations (paper). I estimated the effects of current pollution transport from Asia and Europe on the U.S. and potential climate change impacts on the intercontinental transport. (paper) I continuously evaluated and improved the chemical and physical representations in the global CTM. (paper)

My previous research also had a focus on urban air quality issues. Since 2003, I had been cooperating with the regional climate- air quality (RCAQ) modeling group (led by Dr. Xin-Zhong Liang) in the Illinois State Water Survey (ISWS) to study urban air pollution phenomena and global-regional interactions. Together with Prof. Wuebbles and others, I analyzed potential impacts of changes in regional climate and precursor emissions on ozone pollution over the Northeast U.S. (paper) and the Chicago area, and further compared the different trends of ozone concentrations between the City of Chicago and its surrounding suburban/rural areas. (paper) Working with Dr. Ho-Chun Huang (now in NOAA), I investigated the potential effects of chemical boundary conditions on RCAQ for both current and future years. (paper)

I was also working with Prof. Wuebbles and Dr. Daeok Youn (Now at Seoul National University) on improving the estimation of indirect global warming potential (GWP) for Halons using atmospheric models. (paper)