Estimation of Satellite-Based SO42− and NH4+ Composition of Ambient Fine Particulate Matter over China Using Chemical Transport Model

Abstract

Abstract:Epidemiologic and health impact studies have examined the chemical composition of ambient PM 2.5 in China but have been constrained by the paucity of long-term ground measurements. Using the GEOS-Chem chemical transport model and satellite-derived PM 2.5 data, sulfate and ammonium levels were estimated over China from 2004 to 2014. A comparison of the satellite-estimated dataset with model simulations based on ground measurements obtained from the literature indicated our results are more accurate. Using satellite-derived PM 2.5 data with a spatial resolution of 0.1 × 0.1 • , we further presented finer satellite-estimated sulfate and ammonium concentrations in anthropogenic polluted regions, including the NCP (the North China Plain), the SCB (the Sichuan Basin) and the PRD (the Pearl River Delta). Linear regression results obtained on a national scale yielded an r value of 0.62, NMB of −35.9%, NME of 48.2%, ARB_50% of 53.68% for sulfate and an r value of 0.63, slope of 0.67, and intercept of 5.14 for ammonium. In typical regions, the satellite-derived dataset was significantly robust. Based on the satellite-derived dataset, the spatial-temporal variation of 11-year annual average satellite-derived SO −0.61%; NH 4 + : −0.21%), large values were mainly concentrated in the NCP and SCB. For regions captured at a finer resolution, the inter-annual variation trends presented a positive trend over the periods 2004-2007 and 2008-2011, followed by a negative trend over the period 2012-2014, and sulfate concentrations varied appreciably. Moreover, the seasonal distributions of the 11-year satellite-derived dataset over China were presented. The distribution of both sulfate and ammonium concentrations exhibited seasonal characteristics, with the seasonal concentrations ranking as follows: winter > summer > autumn > spring. High concentrations of these species were concentrated in the NCP and SCB, originating from coal-fired power plants and agricultural activities, respectively. Efforts to reduce sulfur dioxide (SO 2 ) emissions have yielded remarkable results since the government has adopted stricter control measures in recent years. Moreover, ammonia emissions should be controlled while reducing the concentration of sulfur, nitrogen and particulate matter. This study provides an assessment of the population's exposure to certain chemical components.

Separating Daily 1 km PM2.5 Inorganic Chemical Composition in China since 2000 via Deep Learning Integrating Ground, Satellite, and Model Data

“…Two previous estimates of SIAs over China were derived from chemical transport modeling data with composition-specific conversions based on satellite AOD or PM 2.5 estimates but showed poor agreement with in situ measurements (R 2 = 0.38–0.56) for daily SIA estimates of SO 4 2– , NO 3 – , and NH 4 + components at a 10-km resolution. , We also made a comparison with an open-access source, namely, Tracking Air Pollution (TAP) in China (), at the same monitoring stations over the same period (2013–2020), showing an average R 2 of 0.37–0.44 and RMSE of 6.4–10.4 μg/m 3 between daily estimates (10 km) and surface observations of SO 4 2– , NO 3 – , and NH 4 + components (Figure S16). Compared to these studies, our new ChinaHighPMC data set has a ten times higher spatial resolution (1 km) and higher data quality (R 2 = 0.71–0.75 and RMSE = 4.2–6.7 μg/m 3 ) for the three SIA components; our data set also includes another inorganic component, i.e., Cl – .…”

“…This has also been the case in China, where only a handful of studies relying strongly on model simulations have been done. 39,40 What are the concentrations of major species and their proportions to total PM 2.5 mass in China, and what are their temporal changes and impacts on air quality? To address these outstanding questions, we have generated a long-term, daily, seamless PM 2.5 chemical composition product in China at a 1km resolution by applying artificial intelligence to a large ensemble of data sets consisting of ground-based observations of aerosol composition, satellite-derived PM 2.5 at a 1-km resolution, 25 and various auxiliary data, i.e., meteorological reanalyses, pollution emission inventories, and model simulations.…”

“…The long-term evolution of global PM 2.5 components was assessed by integrating satellite PM 2.5 retrievals, GEOS-Chem-simulated PM 2.5 composition information and coincident profiles, and surface PM 2.5 and composition observations. ,− PM 2.5 composition was estimated in other regions with moderate to high resolutions incorporated with surface measurements, such as in southern California at 4.4 km using a generalized additive model, in the northeastern United States at 1 km combining a chemical transport model and geographically weighted regression model or a land-use regression model, and in North America at different resolutions with spatially smoothing models. , However, PM 2.5 composition may have strong spatial gradients due to localized sources and short lifetimes, leading to large estimation uncertainties, especially in regions with sparse observations. This has also been the case in China, where only a handful of studies relying strongly on model simulations have been done. , …”

“…Besides coarse spatial resolutions (0.25°− 0.625°), the chemical transport models did a poorer job in simulating PM 2.5 inorganic components, seriously underestimating the concentrations with stronger deviations, especially for SO4 2− and Cl − (e.g., R 2 ≤ 0.05, slope ≤0.18). By contrast, our estimates improved the spatial resolution drastically by 25−63 times, increasing the correlations by 3−66 times and reducing the RMSE (MAE) values by 36−61% (40− 65%) compared to chemical-model simulations, benefiting from the integration of big data and deep learning.Two previous estimates of SIAs over China were derived from chemical transport modeling data with compositionspecific conversions based on satellite AOD or PM 2.5 estimates but showed poor agreement with in situ measurements (R 2 = 0.38−0.56) for daily SIA estimates of SO 10-km resolution 39,40. We also made a comparison with an open-access source, namely, Tracking Air Pollution (TAP) in China (http://tapdata.org.cn),77 at the same monitoring stations over the same period (2013−2020), showing an average R 2 of 0.37−0.44 and RMSE of 6.4−10.4 μg/m 3 between daily estimates (10 km) and surface observations of SO 4 2−…”

Assessing Effect of Targeting Reduction of PM2.5 Concentration on Human Exposure and Health Burden in Hong Kong Using Satellite Observation

“…Although PM 2.5 concentrations have been regularly monitored at sixteen stations over Hong Kong, these ground monitors still cannot fully cover the entire region. Satellite remote sensing provides an important alternative method toward filling the spatial gap left by fixed-site observations [25][26][27][28]. In this study, we take advantage of high-resolution satellite observations to characterize the spatial variation in PM 2.5 concentration over Hong Kong.…”