PM2.5 Air Pollution Prediction through Deep Learning Using Multisource Meteorological, Wildfire, and Heat Data

被引:12
|
作者
Muthukumar, Pratyush [1 ]
Nagrecha, Kabir [1 ]
Comer, Dawn [2 ]
Calvert, Chisato Fukuda [3 ]
Amini, Navid [1 ]
Holm, Jeanne [2 ]
Pourhomayoun, Mohammad [1 ]
机构
[1] Calif State Univ Los Angeles, Dept Comp Silence, Los Angeles, CA 90032 USA
[2] City Los Angeles, Los Angeles, CA 90012 USA
[3] OpenAQ, Washington, DC 20009 USA
关键词
air pollution prediction; spatiotemporal forecasting; deep convolutional LSTM; remote-sensing satellite imagery; wildfire heat data; meteorological data;
D O I
10.3390/atmos13050822
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Air pollution is a lethal global threat. To mitigate the effects of air pollution, we must first understand it, find its patterns and correlations, and predict it in advance. Air pollution is highly dependent on spatial and temporal correlations of prior meteorological, wildfire, and pollution structures. We use the advanced deep predictive Convolutional LSTM (ConvLSTM) model paired with the cutting-edge Graph Convolutional Network (GCN) architecture to predict spatiotemporal hourly PM2.5 across the Los Angeles area over time. Our deep-learning model does not use atmospheric physics or chemical mechanism data, but rather multisource imagery and sensor data. We use high-resolution remote-sensing satellite imagery from the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument onboard the NASA Terra+Aqua satellites and remote-sensing data from the Tropospheric Monitoring Instrument (TROPOMI), a multispectral imaging spectrometer onboard the Sentinel-5P satellite. We use the highly correlated Fire Radiative Power data product from the MODIS instrument which provides valuable information about the radiant heat output and effects of wildfires on atmospheric air pollutants. The input data we use in our deep-learning model is representative of the major sources of ground-level PM2.5 and thus we can predict hourly PM2.5 at unparalleled accuracies. Our RMSE and NRMSE scores over various site locations and predictive time frames show significant improvement over existing research in predicting PM2.5 using spatiotemporal deep predictive algorithms.
引用
收藏
页数:20
相关论文
共 50 条
  • [1] PM2.5 air pollution prediction through deep learning using meteorological, vehicular, and emission data: A case study of New Delhi, India
    Shakya, Deepti
    Deshpande, Vishal
    Goyal, Manish Kumar
    Agarwal, Mayank
    [J]. JOURNAL OF CLEANER PRODUCTION, 2023, 427
  • [2] PM2.5 CONCENTRATION PREDICTION USING DEEP LEARNING IN AIR MONITORING
    Huang, Yi
    [J]. FRESENIUS ENVIRONMENTAL BULLETIN, 2021, 30 (12): : 13200 - 13211
  • [3] Fine-grained prediction of PM2.5 concentration based on multisource data and deep learning
    Xu, Xiaodi
    Tong, Ting
    Zhang, Wen
    Meng, Lingkui
    [J]. ATMOSPHERIC POLLUTION RESEARCH, 2020, 11 (10) : 1728 - 1737
  • [4] PM2.5 Prediction Based on Random Forest, XGBoost, and Deep Learning Using Multisource Remote Sensing Data
    Joharestani, Mehdi Zamani
    Cao, Chunxiang
    Ni, Xiliang
    Bashir, Barjeece
    Talebiesfandarani, Somayeh
    [J]. ATMOSPHERE, 2019, 10 (07)
  • [5] Ensemble-based deep learning for estimating PM2.5 over California with multisource big data including wildfire smoke
    Li, Lianfa
    Girguis, Mariam
    Lurmann, Frederick
    Pavlovic, Nathan
    McClure, Crystal
    Franklin, Meredith
    Wu, Jun
    Oman, Luke D.
    Breton, Carrie
    Gilliland, Frank
    Habre, Rima
    [J]. ENVIRONMENT INTERNATIONAL, 2020, 145
  • [6] The Effect of Meteorological Conditions on PM10 and PM2.5 Pollution of the Air
    Radzka, Elzbieta
    Rymuza, Katarzyna
    [J]. ROCZNIK OCHRONA SRODOWISKA, 2019, 21 (01): : 611 - 628
  • [7] Predicting PM2.5 atmospheric air pollution using deep learning with meteorological data and ground-based observations and remote-sensing satellite big data
    Pratyush Muthukumar
    Emmanuel Cocom
    Kabir Nagrecha
    Dawn Comer
    Irene Burga
    Jeremy Taub
    Chisato Fukuda Calvert
    Jeanne Holm
    Mohammad Pourhomayoun
    [J]. Air Quality, Atmosphere & Health, 2022, 15 : 1221 - 1234
  • [8] Predicting PM2.5 atmospheric air pollution using deep learning with meteorological data and ground-based observations and remote-sensing satellite big data
    Muthukumar, Pratyush
    Cocom, Emmanuel
    Nagrecha, Kabir
    Comer, Dawn
    Burga, Irene
    Taub, Jeremy
    Calvert, Chisato Fukuda
    Holm, Jeanne
    Pourhomayoun, Mohammad
    [J]. AIR QUALITY ATMOSPHERE AND HEALTH, 2022, 15 (07): : 1221 - 1234
  • [9] Short-term prediction of PM2.5 pollution with deep learning methods
    Ayturan, Y. A.
    Ayturan, Z. C.
    Altun, H. O.
    Kongoli, C.
    Tuncez, F. D.
    Dursun, S.
    Ozturk, A.
    [J]. GLOBAL NEST JOURNAL, 2020, 22 (01): : 126 - 131