Abstract
Abstract. Air pollution is a serious problem in China that urgently needs to be addressed. Air pollution has a great impact on the lives of citizens and on urban development. The particulate matter (PM) value is usually used to indicate the degree of air pollution. In addition to that of PM2.5 and PM10, the use of the PM2.5 ∕ PM10 ratio as an indicator and assessor of air pollution has also become more widespread. This ratio reflects the air pollution conditions and pollution sources. In this paper, a better composite prediction system aimed at improving the accuracy and spatiotemporal applicability of PM2.5 ∕ PM10 was proposed. First, the aerosol optical depth (AOD) in 2017 in Wuhan was obtained based on Moderate Resolution Imaging Spectroradiometer (MODIS) images, with a 1 km spatial resolution, by using the dense dark vegetation (DDV) method. Second, the AOD was corrected by calculating the planetary boundary layer height (PBLH) and relative humidity (RH). Third, the coefficient of determination of the optimal subset selection was used to select the factor with the highest correlation with PM2.5 ∕ PM10 from meteorological factors and gaseous pollutants. Then, PM2.5 ∕ PM10 predictions based on time, space, and random patterns were obtained by using nine factors (the corrected AOD, meteorological data, and gaseous pollutant data) with the long short-term memory (LSTM) neural network method, which is a dynamic model that remembers historical information and applies it to the current output. Finally, the LSTM model prediction results were compared and analyzed with the results of other intelligent models. The results showed that the LSTM model had significant advantages in the average, maximum, and minimum accuracy and the stability of PM2.5 ∕ PM10 prediction.
Highlights
Aerosol is a general term for solid and gas particles suspended in air
The results showed that the average error of the long short-term memory (LSTM) model prediction results is very low, both spatially and temporally, and the stability of the prediction model is significantly better than that of other models
To determine the appropriate number of layers for the LSTM method, except for the data used for prediction, we divided the data set involved in the model construction into three parts: 40 % of the data were used as the training samples for modeling, 30 % of the data were used as the test samples, and the remaining 30 % of the data were used as verification data
Summary
Aerosols can have an important impact on regional and global atmospheric environments, climates, and ecosystems and have long been an important issue in global environmental change research (Crutzen and Andreae, 1990). Particles with an aerodynamic particle size not exceeding 10 μm are called PM10. Particles with an aerodynamic particle size not exceeding 2.5 μm are called fine PM (PM2.5) and are mainly derived from anthropogenic emissions. PM2.5 is mainly produced by anthropogenic combustion for transportation and energy production, and it is important in environmental policy and public health (Xie et al, 2011). Infectious disease research shows that there is a significant consistency between the PM2.5 environmental quality concentration and adverse effects on human health (Lelieveld et al, 2015). Since the scattering extinction contribution of PM2.5 particles accounts for 80 % of the extinction of the atmosphere, the concentration of PM2.5 is a key fac-
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
