Search

Air pollution is a highly concerned environmental issue that have serious impacts on human health and the ecological environment. Accurate air quality prediction can help people effectively deal with the threats posed by air pollution. Most previous work mainly focused on temporal modeling of air quality data at monitoring stations. In recent years, a number of works have used graph convolution to model the spatial dependencies between neighboring sites to extract spatial features and combined temporal model to extract temporal features for predicting PM 2.5. However, these models only considered the local spatial relationships of adjacent sites and ignored sites with longer geographic distance. Therefore, this study proposes a spatio-temporal hybrid model based on convolution and attention (named Attentive Graph Convolution and 1D Convolution Network, AGCC) to predict multi-site and multi-step PM 2.5 concentration. This method not only models the local spatial relationships of adjacent sites, but also combines spatial attention and graph convolutional network (GCN) to model the global spatial relationships of sites. The local and global spatial features are obtained respectively and are fused to obtain spatial features with richer semantic information. Meanwhile, the temporal dependency relationship is modeled through a temporal module composed of temporal attention and one-dimensional convolutional neural network (Conv 1D). AGCC was compared to six baseline models at different prediction horizon based on the air quality datasets of Beijing and Chongqing. The experimental results demonstrate that the model proposed by our achieves the best performance and verifies the feasibility of the model. • The aggregated local and global spatiotemporal information improves the prediction performance. • PM 2.5 prediction of all monitoring stations in Beijing and Chongqing from 1 h to 24 h was realized. • The impact of air quality stations distribution density on PM 2.5 prediction results was investigated. [ABSTRACT FROM AUTHOR]

Cloud is an active component in the weather–climate system that modulates both the radiation balance and the water cycle of the earth system via physical, chemical, and radiative mechanisms. In this study, we used observations of meteorological variables recorded on the Laohugou Glacier No. 12 in the western Qilian Mountains during 2009–2017 to investigate the radiative properties of cloud and its impact on glacier melting. The quantified cloud fraction showed an evident seasonal cycle. The highest cloudiness typically occurred at 16:00 Beijing time, which was probably associated with the strength of local convection that produced frequent occurrence of low-level cumulus or cumulonimbus clouds. Most heavy precipitation events (>14 mm) occurred on overcast days, signified that at least half of the total precipitation could be attributed to transportation from meso- or large-scale atmospheric circulations. Relationships between modelled glacier melting, energy components and cloud fraction showed that clouds could importantly reduce glacier melting, the most important contributor to this process was the clouds impact on net shortwave radiation. Circulation analyses showed that ~7.8%, ~6.3%, and ~ 18.7% of overcast days could be clearly and uniquely attributed to Arctic air mass events, monsoon events, and westerlies events, respectively. The remaining overcast days (~67.2%) were influenced by multiple circulations, e.g., westerlies–monsoon, westerlies–Arctic air mass, monsoon–Arctic air mass, and westerlies–monsoon–Arctic air mass interactions. Monsoon potentially contributes a lot to precipitation in the western Qilian Mountains, future work should aim to do more atmospheric circulation analyses combining with isotopic tracing when precipitation occurs during overcast days. • Clouds and its radiative properties based on in-situ measurements were investigated during 2009–2017. • Occurrence of Clouds explicitly inhibited glacier melting primarily by means of impact on net shortwave radiation. • Overcast condition frequently associated with large-scale circulation. [ABSTRACT FROM AUTHOR]