Assessment of Emission Reduction and Meteorological Change in PM2.5 and Transport Flux in Typical Cities Cluster during 2013–2017

Forecasting Fine Particulate Matter Concentrations by In-Depth Learning Model According to Random Forest and Bilateral Long- and Short-Term Memory Neural Networks

Accurate prediction of fine particulate matter concentration in the future is important for human health due to the necessity of an early warning system. Generally, deep learning methods, when widely used, perform better in forecasting the concentration of PM2.5. However, the source information is limited, and the dynamic process is uncertain. The method of predicting short-term (3 h) and long-term trends has not been achieved. In order to deal with the issue, the research employed a novel mixed forecasting model by coupling the random forest (RF) variable selection and bidirectional long- and short-term memory (BiLSTM) neural net in order to forecast concentrations of PM2.5/0~12 h. Consequently, the average absolute percentage error of 1, 6, and 12 h shows that the PM2.5 concentration prediction is 3.73, 9.33, and 12.68 μg/m3 for Beijing, 1.33, 3.38, and 4.60 μg/m3 for Guangzhou, 1.37, 4.19, and 6.35 μg/m3 for Xi’an, and 2.20, 7.75, and 10.07 μg/m3 for Shenyang, respectively. Moreover, the results show that the suggested mixed model is an advanced method that can offer high accuracy of PM2.5 concentrations from 1 to 12 h post.

The Independent Impacts of PM2.5 Dropping on the Physical and Chemical Properties of Atmosphere over North China Plain in Summer during 2015–2019

Great changes occurred in the physical and chemical properties of the atmosphere in the North China Plain (NCP) in summer caused by PM2.5 dropping from 58 μg/m3 in 2015 to 36.0 μg/m3 in 2019. In this study, we first applied the WRF-Chem model to quantify the impact of PM2.5 reduction on shortwave radiation reaching the ground (SWDOWN), planetary boundary layer height (PBLH), and the surface concentration of air pollutants (represented by CO). Simulation results obtained an increase of 15.0% in daytime SWDOWN and 9.9% in daytime PBLH, and a decrease of −5.0% in daytime CO concentration. These changes were induced by the varied PM2.5 levels. Moreover, the variation in SWDOWN further led to a rise in the NO2 photolysis rate (JNO2) over this region, by 1.82 × 10−4~1.91 × 10−4 s−1 per year. Afterwards, we employed MCM chemical box model to explore how the JNO2 increase and the precursor decrease (CO, VOCs, and NOx) influenced O3 and HOx radicals. The results revealed that the photolysis rate (J) increase would individually cause a change on daytime surface O3, OH, and HO2 radicals by +9.0%, +18.9%, and +23.7%; the corresponding change induced by the precursor decrease was −2.5%, +1.9%, and −2.3%. At the same time, the integrated impacts of the change in J and precursors cause an increase of +6.3%, +21.1%, and +20.9% for daytime surface O3, OH, and HO2. Generally, the atmospheric oxidation capacity significantly enhanced during summer in NCP due to the PM2.5 dropping in recent years. This research can help understand atmosphere changes caused by PM2.5 reduction comprehensively.