期刊
ATMOSPHERIC MEASUREMENT TECHNIQUES
卷 15, 期 3, 页码 721-733出版社
COPERNICUS GESELLSCHAFT MBH
DOI: 10.5194/amt-15-721-2022
关键词
-
资金
- National Centre for Earth Observation [NE/R016518/1]
The study employs a CNN to identify NO2 plumes from satellite data with a success rate of over 90%. Most NO2 plumes are found over mainland China, but there is no correlation with the location of natural gas flaring. Using a CO2 emission inventory, it is determined that the NO2 plume distribution captures 92% of total CO2 emissions.
We use a convolutional neural network (CNN) to identify plumes of nitrogen dioxide (NO2), a tracer of combustion, from NO2 column data collected by the TROPOspheric Monitoring Instrument (TROPOMI). This approach allows us to exploit efficiently the growing volume of satellite data available to characterize Earth's climate. For the purposes of demonstration, we focus on data collected between July 2018 and June 2020. We train the deep learning model using six thousand 28 x 28 pixel images of TROPOMI data (corresponding to similar or equal to 266 km x 133 km) and find that the model can identify plumes with a success rate of more than 90 %. Over our study period, we find over 310 000 individual NO2 plumes, of which similar or equal to 19 % are found over mainland China. We have attempted to remove the influence of open biomass burning using correlative high-resolution thermal infrared data from the Visible Infrared Imaging Radiometer Suite (VIIRS). We relate the remaining NO2 plumes to large urban centres, oil and gas production, and major power plants. We find no correlation between NO2 plumes and the location of natural gas flaring. We also find persistent NO2 plumes from regions where inventories do not currently include emissions. Using an established anthropogenic CO2 emission inventory, we find that our NO2 plume distribution captures 92 % of total CO2 emissions, with the remaining 8 % mostly due to a large number of small sources (< 0.2 g C m(-2) d(-1)) to which our NO2 plume model is less sensitive. We argue that the underlying CNN approach could form the basis of a Bayesian framework to estimate anthropogenic combustion emissions.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据