Adapting a regularized canopy reflectance model (REGFLEC) for the retrieval challenges of dryland agricultural systems

Handle URI:
http://hdl.handle.net/10754/620950
Title:
Adapting a regularized canopy reflectance model (REGFLEC) for the retrieval challenges of dryland agricultural systems
Authors:
Houborg, Rasmus; McCabe, Matthew ( 0000-0002-1279-5272 )
Abstract:
A regularized canopy reflectance model (REGFLEC) is applied over a dryland irrigated agricultural system in Saudi Arabia for the purpose of retrieving leaf area index (LAI) and leaf chlorophyll content (Chll). To improve the robustness of the retrieved properties, REGFLEC was modified to 1) correct for aerosol and adjacency effects, 2) consider foliar dust effects on modeled canopy reflectances, 3) include spectral information in the red-edge wavelength region, and 4) exploit empirical LAI estimates in the model inversion. Using multi-spectral RapidEye imagery allowed Chll to be retrieved with a Mean Absolute Deviation (MAD) of 7.9 μg cm− 2 (16%), based upon in-situ measurements conducted in fields of alfalfa, Rhodes grass and maize over the course of a growing season. LAI and Chll compensation effects on canopy reflectance were largely avoided by informing the inversion process with ancillary LAI inputs established empirically on the basis of a statistical machine learning technique. As a result, LAI was reproduced with good accuracy, with an overall MAD of 0.42 m2 m− 2 (12.5%). Results highlighted the considerable challenges associated with the translation of at-sensor radiance observations to surface bidirectional reflectances in dryland environments, where issues such as high aerosol loadings and large spatial gradients in surface reflectance from bright desert soils to dark vegetated fields are often present. Indeed, surface reflectances in the visible bands were reduced by up to 60% after correction for such adjacency effects. In addition, dust deposition on leaves required explicit modification of the reflectance sub-model to account for its influence. By implementing these model refinements, REGFLEC demonstrated its utility for within-field characterization of vegetation conditions over the challenging landscapes typical of dryland agricultural regions, offering a means through which improvements can be made in the management of these globally important systems.
KAUST Department:
Water Desalination & Reuse Research Cntr; Biological and Environmental Sciences and Engineering (BESE) Division
Citation:
Rasmus Houborg, Matthew F. McCabe, Adapting a regularized canopy reflectance model (REGFLEC) for the retrieval challenges of dryland agricultural systems, Remote Sensing of Environment, Volume 186, 1 December 2016, Pages 105-120, ISSN 0034-4257, http://dx.doi.org/10.1016/j.rse.2016.08.017
Publisher:
Elsevier BV
Journal:
Remote Sensing of Environment
Issue Date:
20-Aug-2016
DOI:
10.1016/j.rse.2016.08.017
Type:
Article
ISSN:
0034-4257
Sponsors:
Research reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST).
Additional Links:
http://www.sciencedirect.com/science/article/pii/S0034425716303200
Appears in Collections:
Articles

Full metadata record

DC FieldValue Language
dc.contributor.authorHouborg, Rasmusen
dc.contributor.authorMcCabe, Matthewen
dc.date.accessioned2016-10-12T10:13:31Z-
dc.date.available2016-10-12T10:13:31Z-
dc.date.issued2016-08-20-
dc.identifier.citationRasmus Houborg, Matthew F. McCabe, Adapting a regularized canopy reflectance model (REGFLEC) for the retrieval challenges of dryland agricultural systems, Remote Sensing of Environment, Volume 186, 1 December 2016, Pages 105-120, ISSN 0034-4257, http://dx.doi.org/10.1016/j.rse.2016.08.017en
dc.identifier.issn0034-4257-
dc.identifier.doi10.1016/j.rse.2016.08.017-
dc.identifier.urihttp://hdl.handle.net/10754/620950-
dc.description.abstractA regularized canopy reflectance model (REGFLEC) is applied over a dryland irrigated agricultural system in Saudi Arabia for the purpose of retrieving leaf area index (LAI) and leaf chlorophyll content (Chll). To improve the robustness of the retrieved properties, REGFLEC was modified to 1) correct for aerosol and adjacency effects, 2) consider foliar dust effects on modeled canopy reflectances, 3) include spectral information in the red-edge wavelength region, and 4) exploit empirical LAI estimates in the model inversion. Using multi-spectral RapidEye imagery allowed Chll to be retrieved with a Mean Absolute Deviation (MAD) of 7.9 μg cm− 2 (16%), based upon in-situ measurements conducted in fields of alfalfa, Rhodes grass and maize over the course of a growing season. LAI and Chll compensation effects on canopy reflectance were largely avoided by informing the inversion process with ancillary LAI inputs established empirically on the basis of a statistical machine learning technique. As a result, LAI was reproduced with good accuracy, with an overall MAD of 0.42 m2 m− 2 (12.5%). Results highlighted the considerable challenges associated with the translation of at-sensor radiance observations to surface bidirectional reflectances in dryland environments, where issues such as high aerosol loadings and large spatial gradients in surface reflectance from bright desert soils to dark vegetated fields are often present. Indeed, surface reflectances in the visible bands were reduced by up to 60% after correction for such adjacency effects. In addition, dust deposition on leaves required explicit modification of the reflectance sub-model to account for its influence. By implementing these model refinements, REGFLEC demonstrated its utility for within-field characterization of vegetation conditions over the challenging landscapes typical of dryland agricultural regions, offering a means through which improvements can be made in the management of these globally important systems.en
dc.description.sponsorshipResearch reported in this publication was supported by funding from King Abdullah University of Science and Technology (KAUST).en
dc.language.isoenen
dc.publisherElsevier BVen
dc.relation.urlhttp://www.sciencedirect.com/science/article/pii/S0034425716303200en
dc.rights© <2016>. This manuscript version is made available under the CC-BY-NC-ND 4.0 license http://creativecommons.org/licenses/by-nc-nd/4.0/en
dc.subjectLAIen
dc.subjectLeaf chlorophyllen
dc.subjectREGFLECen
dc.subjectRapidEyeen
dc.subjectRed-edgeen
dc.subjectAerosolsen
dc.subjectFoliar dusten
dc.subjectAdjacency effectsen
dc.subjectPrecision agricultureen
dc.titleAdapting a regularized canopy reflectance model (REGFLEC) for the retrieval challenges of dryland agricultural systemsen
dc.typeArticleen
dc.contributor.departmentWater Desalination & Reuse Research Cntren
dc.contributor.departmentBiological and Environmental Sciences and Engineering (BESE) Divisionen
dc.identifier.journalRemote Sensing of Environmenten
dc.eprint.versionPost-printen
dc.contributor.affiliationKing Abdullah University of Science and Technology (KAUST)en
kaust.authorHouborg, Rasmusen
kaust.authorMcCabe, Matthewen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.