File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1109/TGRS.2022.3210990
- Scopus: eid_2-s2.0-85139521259
- WOS: WOS:000870339900010
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Estimation of Land Surface Downward Shortwave Radiation Using Spectral-based Convolutional Neural Network Methods: a case study from the Visible Infrared Imaging Radiometer Suite (VIIRS) Images
Title | Estimation of Land Surface Downward Shortwave Radiation Using Spectral-based Convolutional Neural Network Methods: a case study from the Visible Infrared Imaging Radiometer Suite (VIIRS) Images |
---|---|
Authors | |
Keywords | Atmospheric modeling Convolutional Neural Network Convolutional neural networks Deep learning Deep Learning Downward shortwave radiation Earth Land surface Optical reflection Radiative Transfer Remote sensing Spectral-vertical convolution Transfer Learning VIIRS |
Issue Date | 2022 |
Citation | IEEE Transactions on Geoscience and Remote Sensing, 2022 How to Cite? |
Abstract | Surface downward shortwave radiation (DSR) is a key parameter in Earth’s surface radiation budget. Many satellite products have been developed, but their accuracies need further improvements. This study proposed an innovative deep learning method that combines radiative-transfer (RT) modeling with convolutional neural network (CNN) learning for estimating instantaneous DSR from VIIRS observations. Unlike traditional CNN methods that rely on spatial contextual information and are not optimal for medium to coarse resolution satellite data, the proposed algorithm takes advantage of both spectral information as well as vertical information. The algorithm firstly estimates the atmospheric effective optical depth from TOA and surface reflectance by using the look-up table created by radiative transfer simulations. We then constructed a spectral-wised virtual matrix to train the CNN using surface DSR measurements at 34 Baseline Surface Radiation Network sites globally during 2013. The developed CNN was also compared with four traditional machine learning algorithms. The validation results showed that the root mean square error (RMSE) and the bias were 91.42 W/m2 and -0.94 W/m2 respectively. This research is the first spectral-wised CNN application to estimate surface biophysical parameters from satellite remote sensing data quantitively. The comparison with previous look-up table and optimization-based algorithms shows that the proposed algorithm outperforms by around 10~20 W/m2We also explored how transfer learning can further improve the DSR estimation. Our results indicate that the universal model with local data transfer learning outperforms either the CNN with local data or the universal CNN by around 10~20 W/m2. |
Persistent Identifier | http://hdl.handle.net/10722/323166 |
ISSN | 2023 Impact Factor: 7.5 2023 SCImago Journal Rankings: 2.403 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Zhang, Yi | - |
dc.contributor.author | Liang, Shunlin | - |
dc.contributor.author | He, Tao | - |
dc.date.accessioned | 2022-11-18T11:55:10Z | - |
dc.date.available | 2022-11-18T11:55:10Z | - |
dc.date.issued | 2022 | - |
dc.identifier.citation | IEEE Transactions on Geoscience and Remote Sensing, 2022 | - |
dc.identifier.issn | 0196-2892 | - |
dc.identifier.uri | http://hdl.handle.net/10722/323166 | - |
dc.description.abstract | Surface downward shortwave radiation (DSR) is a key parameter in Earth’s surface radiation budget. Many satellite products have been developed, but their accuracies need further improvements. This study proposed an innovative deep learning method that combines radiative-transfer (RT) modeling with convolutional neural network (CNN) learning for estimating instantaneous DSR from VIIRS observations. Unlike traditional CNN methods that rely on spatial contextual information and are not optimal for medium to coarse resolution satellite data, the proposed algorithm takes advantage of both spectral information as well as vertical information. The algorithm firstly estimates the atmospheric effective optical depth from TOA and surface reflectance by using the look-up table created by radiative transfer simulations. We then constructed a spectral-wised virtual matrix to train the CNN using surface DSR measurements at 34 Baseline Surface Radiation Network sites globally during 2013. The developed CNN was also compared with four traditional machine learning algorithms. The validation results showed that the root mean square error (RMSE) and the bias were 91.42 W/m2 and -0.94 W/m2 respectively. This research is the first spectral-wised CNN application to estimate surface biophysical parameters from satellite remote sensing data quantitively. The comparison with previous look-up table and optimization-based algorithms shows that the proposed algorithm outperforms by around 10~20 W/m2We also explored how transfer learning can further improve the DSR estimation. Our results indicate that the universal model with local data transfer learning outperforms either the CNN with local data or the universal CNN by around 10~20 W/m2. | - |
dc.language | eng | - |
dc.relation.ispartof | IEEE Transactions on Geoscience and Remote Sensing | - |
dc.subject | Atmospheric modeling | - |
dc.subject | Convolutional Neural Network | - |
dc.subject | Convolutional neural networks | - |
dc.subject | Deep learning | - |
dc.subject | Deep Learning | - |
dc.subject | Downward shortwave radiation | - |
dc.subject | Earth | - |
dc.subject | Land surface | - |
dc.subject | Optical reflection | - |
dc.subject | Radiative Transfer | - |
dc.subject | Remote sensing | - |
dc.subject | Spectral-vertical convolution | - |
dc.subject | Transfer Learning | - |
dc.subject | VIIRS | - |
dc.title | Estimation of Land Surface Downward Shortwave Radiation Using Spectral-based Convolutional Neural Network Methods: a case study from the Visible Infrared Imaging Radiometer Suite (VIIRS) Images | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1109/TGRS.2022.3210990 | - |
dc.identifier.scopus | eid_2-s2.0-85139521259 | - |
dc.identifier.eissn | 1558-0644 | - |
dc.identifier.isi | WOS:000870339900010 | - |