File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: A machine learning method trained by radiative transfer model inversion for generating seven global land and atmospheric estimates from VIIRS top-of-atmosphere observations

TitleA machine learning method trained by radiative transfer model inversion for generating seven global land and atmospheric estimates from VIIRS top-of-atmosphere observations
Authors
KeywordsMachine learning
Multiple variables
Radiative transfer models (RTMs)
Random forest regression
VIIRS product
Issue Date2022
Citation
Remote Sensing of Environment, 2022, v. 279, article no. 113132 How to Cite?
AbstractThe Visible Infrared Imaging Radiometer Suite (VIIRS) has observed the Earth since 2011 and will continue for several decades. Unfortunately, few high-level land surface products have been produced and publicly released. The numerical inversion of radiative transfer model (RTM) has long been used for the retrieval of land surface and atmospheric variables from satellite data; however, it is computationally intensive. Some methods (e.g., using look-up tables) are efficient but are most suitable for estimating individual variable. Various machine learning (ML) models have been developed, mostly trained by either ground measurements that may not be adequate to represent different conditions, or RTM simulations that may not be realistic due to model uncertainty and unnecessary combinations of illumination-viewing geometries and variable values. In this study, we replaced the numerical inversion scheme by the ML algorithm that is trained using RTM inversion for estimating multiple variables simultaneously. Our training dataset for the ML models was generated by inverting the land and atmospheric variables from VIIRS top-of-atmosphere (TOA) reflectance data using a coupled land-surface–atmosphere RT model. Two multi-output ML algorithms were explored: backpropagation neural networks and random forest (RF) regression. The best-performing RF model was then applied to estimate seven land and atmospheric variables globally from VIIRS TOA data: the leaf area index (LAI), incident photosynthetically active radiation (PAR), fraction of absorbed photosynthetically active radiation (FAPAR), incident shortwave radiation (ISR), land surface albedo, land surface reflectance, and TOA albedo. The inversion results were validated using ground measurements at 54 sites and showed comparable accuracy to the numerical inversion scheme that is suitable for only small regions. An experiment was also conducted to test this method on a much larger region for detailed evaluation. Finally, seven global estimates in 2013 were experimentally produced. Inter-comparison with the existing satellite products showed that the VIIRS estimates had high consistency with the LAI and FAPAR products of the Global LAnd Surface Satellite (GLASS) products suite. Moreover, the VIIRS officially released the VNP43MA3 albedo and VNP09A1 surface reflectance products, the retrieved estimates also matched them well. It was also shown that simultaneous estimation of all seven variables was more accurate and efficient than estimating individual variable sequentially.
Persistent Identifierhttp://hdl.handle.net/10722/316663
ISSN
2023 Impact Factor: 11.1
2023 SCImago Journal Rankings: 4.310
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorZhang, Guodong-
dc.contributor.authorMa, Han-
dc.contributor.authorLiang, Shunlin-
dc.contributor.authorJia, Aolin-
dc.contributor.authorHe, Tao-
dc.contributor.authorWang, Dongdong-
dc.date.accessioned2022-09-14T11:41:00Z-
dc.date.available2022-09-14T11:41:00Z-
dc.date.issued2022-
dc.identifier.citationRemote Sensing of Environment, 2022, v. 279, article no. 113132-
dc.identifier.issn0034-4257-
dc.identifier.urihttp://hdl.handle.net/10722/316663-
dc.description.abstractThe Visible Infrared Imaging Radiometer Suite (VIIRS) has observed the Earth since 2011 and will continue for several decades. Unfortunately, few high-level land surface products have been produced and publicly released. The numerical inversion of radiative transfer model (RTM) has long been used for the retrieval of land surface and atmospheric variables from satellite data; however, it is computationally intensive. Some methods (e.g., using look-up tables) are efficient but are most suitable for estimating individual variable. Various machine learning (ML) models have been developed, mostly trained by either ground measurements that may not be adequate to represent different conditions, or RTM simulations that may not be realistic due to model uncertainty and unnecessary combinations of illumination-viewing geometries and variable values. In this study, we replaced the numerical inversion scheme by the ML algorithm that is trained using RTM inversion for estimating multiple variables simultaneously. Our training dataset for the ML models was generated by inverting the land and atmospheric variables from VIIRS top-of-atmosphere (TOA) reflectance data using a coupled land-surface–atmosphere RT model. Two multi-output ML algorithms were explored: backpropagation neural networks and random forest (RF) regression. The best-performing RF model was then applied to estimate seven land and atmospheric variables globally from VIIRS TOA data: the leaf area index (LAI), incident photosynthetically active radiation (PAR), fraction of absorbed photosynthetically active radiation (FAPAR), incident shortwave radiation (ISR), land surface albedo, land surface reflectance, and TOA albedo. The inversion results were validated using ground measurements at 54 sites and showed comparable accuracy to the numerical inversion scheme that is suitable for only small regions. An experiment was also conducted to test this method on a much larger region for detailed evaluation. Finally, seven global estimates in 2013 were experimentally produced. Inter-comparison with the existing satellite products showed that the VIIRS estimates had high consistency with the LAI and FAPAR products of the Global LAnd Surface Satellite (GLASS) products suite. Moreover, the VIIRS officially released the VNP43MA3 albedo and VNP09A1 surface reflectance products, the retrieved estimates also matched them well. It was also shown that simultaneous estimation of all seven variables was more accurate and efficient than estimating individual variable sequentially.-
dc.languageeng-
dc.relation.ispartofRemote Sensing of Environment-
dc.subjectMachine learning-
dc.subjectMultiple variables-
dc.subjectRadiative transfer models (RTMs)-
dc.subjectRandom forest regression-
dc.subjectVIIRS product-
dc.titleA machine learning method trained by radiative transfer model inversion for generating seven global land and atmospheric estimates from VIIRS top-of-atmosphere observations-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.rse.2022.113132-
dc.identifier.scopuseid_2-s2.0-85132720955-
dc.identifier.volume279-
dc.identifier.spagearticle no. 113132-
dc.identifier.epagearticle no. 113132-
dc.identifier.isiWOS:000841468600001-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats