Evaluating the Governing Factors of Variability in Nocturnal Boundary Layer Height Based on Elastic Lidar in Wuhan

Liu

et al. 2021

Atmos. Chem. Phys.

Self Cite

Abstract. The vertical distribution of aerosol extinction coefficient (EC) measured by lidar systems has been used to retrieve the profile of particle matter with a diameter <2.5 µm (PM2.5). However, the traditional linear model (LM) cannot consider the influence of multiple meteorological variables sufficiently and then induce the low inversion accuracy. Generally, the machine learning (ML) algorithms can input multiple features which may provide us with a new way to solve this constraint. In this study, the surface aerosol EC and meteorological data from January 2014 to December 2017 were used to explore the conversion of aerosol EC to PM2.5 concentrations. Four ML algorithms were used to train the PM2.5 prediction models: random forest (RF), K-nearest neighbor (KNN), support vector machine (SVM) and extreme gradient boosting decision tree (XGB). The mean absolute error (root mean square error) of LM, RF, KNN, SVM and XGB models were 11.66 (15.68), 5.35 (7.96), 7.95 (11.54), 6.96 (11.18) and 5.62 (8.27) µg/m3, respectively. This result shows that the RF model is the most suitable model for PM2.5 inversions from EC and meteorological data. Moreover, the sensitivity analysis of model input parameters was also conducted. All these results further indicated that it is necessary to consider the effect of meteorological variables when using EC to retrieve PM2.5 concentrations. Finally, the diurnal and seasonal variations of transport flux (TF) and PM2.5 profiles were analyzed based on the lidar data. The large PM2.5 concentration occurred at approximately 13:00–17:00 local time (LT) in 0.2–0.8 km. The diurnal variations of the TF show a clear conveyor belt at approximately 12:00–18:00 LT in 0.5–0.8 km. The results indicated that air pollutant transport over Wuhan mainly occurs at approximately 12:00–18:00 LT in 0.5–0.8 km. The TF near the ground usually has the highest value in winter (0.26 mg/m2 s), followed by the autumn and summer (0.2 and 0.19 mg/m2 s, respectively), and the lowest value in spring (0.14 mg/m2 s). These findings give us important information on the atmospheric profile and provide us sufficient confidence to apply lidar in the study of air quality monitoring.

Section: Vertical Evolution Of Pm 25 and Tfsupporting

confidence: 92%

Estimation of the vertical distribution of particle matter (PM<sub>2.5</sub>) concentration and its transport flux from lidar measurements based on machine learning algorithms

Liu

et al. 2021

Atmos. Chem. Phys.

Self Cite

“…The concentration of PM2.5 at 0.2 km has the highest value in the winter (93.7 ug/m 3 ), followed by the autumn and summer (80.3 and 75.8 ug/m 3 , respectively), and lowest in the spring (53.5 ug/m 3 ). This finding is similar to the surface observation results (Wang et al, 2016). The PM2.5 concentration decreases gradually with the height increases.…”

Section: Vertical Evolution Of Pm25 and Tfsupporting

confidence: 92%

Estimation of the vertical distribution of particle matter (PM2.5) concentration and its transport flux from lidar measurements based on machine learning algorithms

et al. 2021

Preprint

Self Cite

Abstract. The vertical distribution of aerosol extinction coefficient (EC) measured by lidar system has been used to retrieve the profile of particle matter with a diameter < 2.5 μm (PM2.5). However, the traditional linear model (LM) cannot consider the influence of multiple meteorological variables sufficiently, and then inducing the low inversion accuracy. Generally, the machine learning (ML) algorithms can input multiple features which may provide us with a new way to solve this constraint. In this study, the surface aerosol EC and meteorological data from January 2014 to December 2017 were used to explore the conversion of aerosol EC to PM2.5 concentrations. Four ML algorithms were used to train the PM2.5 prediction models, including Random Forest (RF), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and eXtreme Gradient Boosting Decision Tree (XGB). The mean absolute error (root mean square error) of LM, RF, KNN, SVM and XGB models were 11.66 (15.68), 5.35 (7.96), 7.95 (11.54), 6.96 (11.18) and 5.62 (8.27) μg/m3, respectively. This result show that the RF model is the most suitable model for PM2.5 inversions from EC and meteorological data. Moreover, the sensitivity analysis of model input parameters was also conducted. All these results further indicated that it is necessary to consider the effect of meteorological variables when using EC to retrieve PM2.5 concentrations. Finally, the diurnal and seasonal variations of transport flux (TF) and PM2.5 profiles were analyzed based on the lidar data. The large PM2.5 concentration occurred at approximately 13:00–17:00 Location Time (LT) in 0.2–0.8 km. The diurnal variations of the TF shows a clear conveyor belt at approximately 12:00–18:00 LT in 0.5–0.8 km. These results indicated that air pollutants transport over Wuhan mainly occurs at approximately 12:00–18:00 LT in 0.5–0.8 km. The TF near the ground usually have the highest value in winter (0.26 mg/m2 s), followed by the autumn and summer (0.2 and 0.19 mg/m2 s), respectively, and the lowest value in spring (0.14 mg/m2 s). These findings give us important information of atmospheric profile and provide us sufficient confidence to apply lidar in the study of air quality monitoring.

“…What is more, numerous epidemiological studies have indicated that the suspended particles in the atmosphere, especially those with an aerodynamic diameter not larger than 2.5 µm (PM 2.5 ), would produce serious adverse effects on human bodies, hence increasing deaths from cardiovascular and respiratory diseases and even lung cancer [7,8]. As such, aerosols have drawn considerable concern [9,10,11]. In recent years, numerous studies have been conducted to investigate the formation mechanisms of aerosol pollution, and some of them implied the important role of relative humidity ( RH ) in this issue because of the aerosol hygroscopicity [12,13].…”

Section: Introductionmentioning

confidence: 99%

Roles of Relative Humidity in Aerosol Pollution Aggravation over Central China during Wintertime

Zang

Wang

et al. 2019

IJERPH

Aerosol pollution elicits considerable public concern due to the adverse influence on air quality, climate change, and human health. Outside of emissions, haze formation is closely related to meteorological conditions, especially relative humidity (RH). Partly due to insufficient investigations on the aerosol hygroscopicity, the accuracy of pollution prediction in Central China is limited. In this study, taking Wuhan as a sample city, we investigated the response of aerosol pollution to RH during wintertime based on in-situ measurements. The results show that, aerosol pollution in Wuhan is dominated by PM2.5 (aerodynamic particle size not larger than 2.5 μm) on wet days (RH ≥ 60%), with the averaged mass fraction of 0.62 for PM10. Based on the RH dependence of aerosol light scattering (f (RH)), aerosol hygroscopicity was evaluated and shows the high dependence on the particle size distribution and chemical compositions. f (RH = 80%) in Wuhan was 2.18 (±0.73), which is comparable to that measured in the Pearl River Delta and Yangtze River Delta regions for urban aerosols, and generally greater than values in Beijing. Ammonium (NH4+), sulfate (SO42−), and nitrate (NO3−) were enhanced by approximately 2.5-, 2-, and 1.5-fold respectively under wet conditions, and the ammonia-rich conditions in wintertime efficiently promoted the formation of SO42− and NO3−, especially at high RH. These secondary ions play an important role in aggravating the pollution level and aerosol light scattering. This study has important implications for understanding the roles of RH in aerosol pollution aggravation over Central China, and the fitted equation between f (RH) and RH may be helpful for pollution forecasting in this region.