Spatial and seasonal variations of aerosols over China from two decades of multi-satellite observations – Part 2: AOD time series for 1995–2017 combined from ATSR ADV and MODIS C6.1 and AOD tendency estimations

Abstract

Abstract. Understanding long-term variations in aerosol loading is essential for evaluating the health and climate effects of airborne particulates as well as the effectiveness of pollution control policies. The expected satellite lifetime is about 10 to 15 years. Therefore, to study the variations of atmospheric constituents over longer periods information from different satellites must be utilized. Here we introduce a method to construct a combined annual and seasonal long time series of AOD at 550 nm using the Along-Track Scanning Radiometers (ATSR: ATSR-2 and AATSR combined) and the MODerate resolution Imaging Spectroradiometer on Terra (MODIS/Terra), which together cover the 1995–2017 period. The long-term (1995–2017) combined AOD time series are presented for all of mainland China, for southeastern (SE) China and for 10 selected regions in China. Linear regression was applied to the combined AOD time series constructed for individual L3 (1∘ × 1∘) pixels to estimate the AOD tendencies for two periods: 1995–2006 (P1) and 2011–2017 (P2), with respect to the changes in the emission reduction policies in China. During P1, the annually averaged AOD increased by 0.006 (or 2 % of the AOD averaged over the corresponding period) per year across all of mainland China, reflecting increasing emissions due to rapid economic development. In SE China, the annual AOD positive tendency in 1995–2006 was 0.014 (3 %) per year, reaching maxima (0.020, or 4 %, per year) in Shanghai and the Pearl River Delta regions. After 2011, during P2, AOD tendencies reversed across most of China with the annually averaged AOD decreasing by −0.015 (−6 %) per year in response to the effective reduction of the anthropogenic emissions of primary aerosols, SO2 and NOx. The strongest AOD decreases were observed in the Chengdu (−0.045, or −8 %, per year) and Zhengzhou (−0.046, or −9 %, per year) areas, while over the North China Plain and coastal areas the AOD decrease was lower than −0.03 (approximately −6 %) per year. In the less populated areas the AOD decrease was small. The AOD tendency varied by both season and region. The increase in the annually averaged AOD during P1 was mainly due to an increase in summer and autumn in SE China (0.020, or 4 %, and 0.016, or 4 %, per year, respectively), while during winter and spring the AOD actually decreased over most of China. The AOD negative tendencies during the 2011–2017 period were larger in summer than in other seasons over the whole of China (ca. −0.021, or −7 %, per year) and over SE China (ca. −0.048, or −9 %, per year). The long-term AOD variations presented here show a gradual decrease in the AOD after 2011 with an average reduction of 30 %–50 % between 2011 and 2017. The effect is more visible in the highly populated and industrialized regions in SE China, as expected.

Interannual variability and trends of combustion aerosol and dust in major continental outflows revealed by MODIS retrievals and CAM5 simulations during 2003–2017

“…Alfaro-Contreras et al (2017) analyzed several satellite retrievals of AOD for total aerosol and found no consistent trend in coastal China (−0.035, +0.001, −0.010 per decade for MODIS Aqua, MODIS Terra, and MISR, respectively). Sogacheva et al (2018) combined AOD data from the Along-Track Scanning Radiometer (ATSR) and MODIS and found that AOD across China increased significantly from 1995 to 2006 but decreased gradually between 2011 and 2017, which reflects the increased emissions due to rapid economic development and the decreased emissions due to effective emission control regulations. For com-parison, the CAM5 simulations show no statistically significant trend.…”

“…Our results are also consistent with trends of dust emissions over East Asia and China in particular as documented in the literature. For example, Song et al (2016) showed that the spring dust storm frequency in arid and semiarid regions of China has decreased by 15.45 storms per year on average over the period of 1982 to 2007. Fan et al (2014) showed that the decrease in springtime dust storms in Inner Mongolia, northern China, from 1982 to 2008 was correlated with advanced vegetation growth, with 1 d earlier green-up data corresponding to a 3 % decrease in spring dust storms.…”

Long-Term (2005–2017) View of Atmospheric Pollutants in Central China Using Multiple Satellite Observations

“…Thus, despite the spatial inconsistency, the substantial variability of SO2 and NOx in central China can cause a considerable contribution to overall variations of overall aerosol loading. Spatial mean of aerosol loading is usually used to evaluate variations of regional emission levels [34]. The annual average of MODIS AOD values within Hubei province during 2005-2017 are shown in Figure 10.…”

Investigations into the development of a satellite-based aerosol climate data record using ATSR-2, AATSR and AVHRR data over north-eastern China from 1987 to 2012

“…Major problems in the retrieval of AOD from satellite observations are the effective decoupling of atmospheric and surface effects on the reflectance measured at the TOA, cloud detection, and the description of the aerosol properties. In the ADL algorithm a cloud mask is applied based on the clouds from AVHRR (CLAVR, Stowe et al, 1991Stowe et al, , 1999, and in the retrieval six aerosol types are used as proposed by Govaerts (2010). For cloud-free pixels, and after application of gas absorption corrections as described in Xue et al (2017), the land surface reflectance in the AVHRR channel 1 (0.64 µm) is parameterised in terms of the measured reflectance in channel 3 (3.75 µm), with coefficients which are functions of the NDVI and scattering angle.…”