3D radiative transfer modelling of fire impacts on a two-layer savanna system

We present a new, detailed three dimensional (3D) approach to modelling the pre- and post-fire reflectance of a two-layer savanna system modelled as heterogeneous overstory (tree) and understory (grass) layers. The models were developed from detailed field measurements of structural and radiometric...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Remote sensing of environment 2011-08, Vol.115 (8), p.1866-1881
Hauptverfasser: Disney, M.I., Lewis, P., Gomez-Dans, J., Roy, D., Wooster, M.J., Lajas, D.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:We present a new, detailed three dimensional (3D) approach to modelling the pre- and post-fire reflectance of a two-layer savanna system modelled as heterogeneous overstory (tree) and understory (grass) layers. The models were developed from detailed field measurements of structural and radiometric properties made at experimental burn plots with varying canopy cover in the Kruger National Park, South Africa. The models were used to simulate 400–2500 nm spectral reflectance at 10–500 m spatial scale for various viewing and solar geometry configurations. The model simulations closely matched pre-fire and post-fire ground-based, helicopter and satellite remote sensing observations (all r 2 values > 0.95 except one post-fire case). The largest discrepancies between modelled and observed reflectances occurred typically at wavelengths greater than 1200 nm for the post-fire simulations. The modelling results indicate that representation of overstory and understory structure and scattering properties are required to represent the burn signal in a typical savanna system. The described 3D modelling approach enables separation of the scattering contributions of the different scene components and is suited to testing and validating fire impact assessment algorithms at locations where the difficulty of obtaining both pre- and post-fire observations is a severe constraint. ► We present 3D models of savanna systems, including detailed tree and grass structure. ► The models describe and explain the pre- and post-fire reflectance signal of a savanna system. ► The models describe reflectance very well over a range of scales 100 m. ► We discuss the use of such models to test simple models of fire effects and develop new algorithms.
ISSN:0034-4257
1879-0704
DOI:10.1016/j.rse.2011.03.010