KIT | KIT-Bibliothek | Impressum | Datenschutz

High-Resolution PM$_{10}$ Estimation Using Satellite Data and Model-Agnostic Meta-Learning

Yang, Yue ; Cermak, Jan ORCID iD icon 1,2; Chen, Xu; Chen, Yunping; Hou, Xi
1 Institut für Meteorologie und Klimaforschung – Atmosphärische Spurenstoffe und Fernerkundung (IMK-ASF), Karlsruher Institut für Technologie (KIT)
2 Institut für Photogrammetrie und Fernerkundung (IPF), Karlsruher Institut für Technologie (KIT)

Abstract:

Characterizing the spatial distribution of particles smaller than 10 μm (PM$_{10}$) is of great importance for air quality management yet is very challenging because of the sparseness of air quality monitoring stations. In this study, we use a model-agnostic meta-learning-trained artificial neural network (MAML-ANN) to estimate the concentrations of PM$_{10}$ at 60 m × 60 m spatial resolution by combining satellite-derived aerosol optical depth (AOD) with meteorological data. The network is designed to regress from the predictors at a specific time to the ground-level PM$_{10}$ concentration. We utilize the ANN model to capture the time-specific nonlinearity among aerosols, meteorological conditions, and PM$_{10}$, and apply MAML to enable the model to learn the nonlinearity across time from only a small number of data samples. MAML is also employed to transfer the knowledge learned from coarse spatial resolution to high spatial resolution. The MAML-ANN model is shown to accurately estimate high-resolution PM$_{10}$ in Beijing, with coefficient of determination of 0.75. MAML improves the PM$_{10}$ estimation performance of the ANN model compared with the baseline using pre-trained initial weights. ... mehr


Verlagsausgabe §
DOI: 10.5445/IR/1000172925
Veröffentlicht am 19.08.2024
Cover der Publikation
Zugehörige Institution(en) am KIT Institut für Meteorologie und Klimaforschung – Atmosphärische Spurenstoffe und Fernerkundung (IMK-ASF)
Institut für Photogrammetrie und Fernerkundung (IPF)
Publikationstyp Zeitschriftenaufsatz
Publikationsjahr 2024
Sprache Englisch
Identifikator ISSN: 2072-4292
KITopen-ID: 1000172925
Erschienen in Remote Sensing
Verlag MDPI
Band 16
Heft 13
Seiten Art.-Nr.: 2498
Vorab online veröffentlicht am 08.07.2024
Nachgewiesen in Dimensions
Web of Science
Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page