[Characterization of Atmospheric Composite Pollution Prediction and Dynamic Cross-transmission in the Beijing-Tianjin-Hebei Urban Agglomeration].

环境科学 污染 污染物 气象学 空气污染 北京 风速 大气科学 大气污染 百分位 连接词(语言学) 环境工程 集聚经济
作者
Hongmin Li,Zhong-Qiu Ren
出处
期刊:PubMed [National Institutes of Health]
卷期号:47 (1): 210-222 被引量:2
标识
DOI:10.13227/j.hjkx.202412119
摘要

The dynamic cross-transmission characteristics of air pollution between cities were explored based on the data of AQI, PM2.5, and other important pollutant concentrations and meteorological indicators in the Beijing-Tianjin-Hebei urban agglomeration from 2017 to 2023, using a Copula function-based feature selection model to identify the key features of composite atmospheric pollution, constructing a Bayesian Extreme Gradient Boosting (BOA-XGBoost) model for dynamic prediction of composite atmospheric pollution in urban agglomeration, and combining complex network analysis and SHAP interpretable modeling. The results show that: ① The feature selection model based on Copula function successfully identified the atmospheric pollutants that were highly correlated with the AQI as PM2.5, PM10, O3, and NO2 and meteorological factors as the average temperature and average humidity.② Compared with other models, the BOA-XGBoost model performed optimally under the comprehensive evaluation system, in which the R2, MAE, RMSE, and MAPE of the Beijing test set reached 0.970 4, 3.965 5, 8.351 8, and 0.062 4, respectively. ③ The pollution transmission network of the Beijing-Tianjin-Hebei urban agglomeration has gradually evolved into a "core-periphery" model, with Tianjin and Cangzhou as the core nodes of pollution transmission weakening and Zhangjiakou, Chengde, and Qinhuangdao being in the periphery for a long time. ④ There was heterogeneity in the strength of pollution transport between urban subgroups, but pollution transport was weakened within them. Pollutant contributions showed that PM2.5 and PM10 dominated the AQI changes in the subgroups, followed by O3, while other pollutants had small and complex effects. ⑤ The city cluster mainly consisted of "double-high" and "double-low" cities, with Beijing switching from "double-high" to "sink" in 2020. The "double-high" cities such as Tianjin and Shijiazhuang will remain the center of pollution transmission.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
秋风应助科研通管家采纳,获得10
刚刚
刚刚
英俊的铭应助科研通管家采纳,获得10
刚刚
刚刚
科研通AI2S应助科研通管家采纳,获得10
刚刚
刚刚
刚刚
刚刚
无花果应助科研通管家采纳,获得10
刚刚
刚刚
CodeCraft应助科研通管家采纳,获得10
1秒前
1秒前
Joanna完成签到 ,获得积分10
1秒前
猪幺妖完成签到 ,获得积分10
4秒前
4秒前
科研通AI6.2应助tudou采纳,获得10
4秒前
4秒前
611发布了新的文献求助10
5秒前
顾矜应助xsy采纳,获得10
6秒前
taipingyang发布了新的文献求助10
6秒前
8秒前
9秒前
swh发布了新的文献求助10
9秒前
谢雷XIELei应助sssugar采纳,获得10
9秒前
时尚的秋天完成签到 ,获得积分10
11秒前
传奇3应助积极障碍患者采纳,获得10
11秒前
李健的小迷弟应助比迪奇采纳,获得10
12秒前
FeLaN发布了新的文献求助10
13秒前
611完成签到,获得积分10
15秒前
15秒前
RC完成签到 ,获得积分10
16秒前
缥缈冥发布了新的文献求助10
16秒前
Djtc发布了新的文献求助20
16秒前
16秒前
16秒前
18秒前
焚琴涮羊肉完成签到,获得积分10
18秒前
18秒前
18秒前
JamesPei应助李彬采纳,获得10
19秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
Encyclopedia of Cardiovascular Research and Medicine(2e) 820
自動車の空力技術 800
Essentials of Carbohydrate Chemistry and Biochemistry, 4th Edition 800
Organizational Behavior 510
Management and the Arts 510
Matrix Methods in Data Mining and Pattern Recognition Second Edition 510
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 化学工程 工程类 有机化学 物理 复合材料 生物化学 内科学 细胞生物学 基因 遗传学 免疫学 冶金 光电子学 癌症研究
热门帖子
关注 科研通微信公众号,转发送积分 7781701
求助须知:如何正确求助?哪些是违规求助? 9321323
关于积分的说明 20382462
捐赠科研通 7369503
什么是DOI,文献DOI怎么找? 3320074
关于科研通互助平台的介绍 2467923
邀请新用户注册赠送积分活动 2336029