Using Air Quality Models and Satellite Observations to Assess Pollution in India

Ambient air pollution of particulate and gaseous constituents is a persistent health problem globally, contributing to nearly 3.7 million premature deaths each year. In India, high concentrations of fine particulate matter (PM_2.5) occur regularly, several Indian cities consistently rank among the world’s most polluted cities, and an estimated 580 thousand deaths every year are related to high levels of PM_2.5 and ozone (O₃) pollution. Pollution in India is dominated by energy sector emissions of primary particulates, sulfur dioxide (SO₂), and nitrogen dioxide (NO₂), and sources include stationary sources like electricity generation and industrial manufacturing and processing, mobile sources, and domestic combustion. Emission sectors impact ambient air quality differently due to spatial distribution (typical urban vs. typical rural sources) as well as source height characteristics (low-level vs. high stack sources). Air quality monitoring by the Central Pollution Control Board (CPCB) in India is mostly limited to urban areas, and resulting concentrations are used in health impact assessments, meaning health impacts on India’s 1.25 billion people may actually be underestimated. Therefore, air quality analysis tools with substantial spatial and temporal coverage are important for more detailed assessment of air quality in the region.

To support use of a policy-relevant regional air quality model for analysis in India, we use satellite and surface observations to evaluate simulations of four seasonally representative months from the U.S. Environmental Protection Agency (EPA) Community Multi-Scale Air Quality (CMAQ) model. Simulations are conducted for January, April, July, and October 2010 at 36km by 36km and vertically through 36 layers in the troposphere. Inputs include biogenic emissions from the Community Land Model coupled with the Model of Emissions of Gases and Aerosols in Nature (CLM-MEGAN), biomass-burning emissions from the Global Fires Emissions Database (GFED), and ERA-Interim meteorology generated with the Weather Research and Forecasting (WRF) model. Annual anthropogenic emissions from the Greenhouse Gas-Air Pollution Interaction and Synergies (GAINS) model (ECLIPSE v5a) are regridded from 0.5° by 0.5° to 36 km by 36 km and vertically distributed depending on sector. Model evaluation is conducted using satellite tropospheric column observations of NO₂ from the Ozone Monitoring Instrument (OMI). Additional model evaluation for NO₂, SO₂, O₃, and PM­_2.5 is conducted using surface observations at monitor locations across India from the CPCB and from the peer-reviewed literature. Evaluation with satellite observations indicate model low biases in the tropospheric column (-65.8%), and evaluation with surface observations indicate model high biases, particularly in urban areas, for gas-phase species (NO₂: 31.4%, SO₂: 26.7%, O₃: 19.7%) and low model biases for fine particulates (PM_2.5: -47.1%). Results have implication for air quality model support in decision-making to improve air quality and health impacts in India.