FluorWPS: A Monte Carlo ray-tracing model to compute sun-induced chlorophyll fluorescence of three-dimensional canopy

REMOTE SENSING OF ENVIRONMENT

卷:187

Article

Zhao, Feng¹  Dai, Xu¹  Verhoef, Wout²  Guo, Yiqing^1,3  van der Tol, Christiaan²  Li, Yuguang^1,4  Huang, Yanbo⁵ 

[1] Beihang Univ, Sch Instrumentat Sci & Optoelect Engn, Beijing 100191, Peoples R China

[2] Univ Twente, Fac Geoinformat Sci & Earth Observat ITC, POB 217, NL-7500 AE Enschede, Netherlands

[3] Univ New South Wales, Sch Engn & Informat Technol, Canberra, ACT 2600, Australia

[4] Univ Washington, Dept Elect Engn, 185 Stevens Way, Seattle, WA 98195 USA

[5] ARS, USDA, Crop Reduct Syst Res Unit, 141 Expt Stn Rd, Stoneville, MS 38776 USA

关键词: Sun-induced chlorophyll fluorescence (SIF);    Radiative transfer;    Monte Carlo model;    Fluorescence model with weighted photon;    spread method (FluorWPS);   

DOI  :  10.1016/j.rse.2016.10.036

来源: Elsevier

【 摘 要 】

A model to simulate radiative transfer (RT) of sun-induced chlorophyll fluorescence (SIF) of three-dimensional (3-D) canopy, FluorWPS (Fluorescence model with Weighted Photon Spread method), was proposed and evaluated. The inclusion of fluorescence excitation was implemented with the 'weight reduction' and 'photon spread' concepts based on Monte Carlo ray-tracing technique. The radiative transfer of SIF in a 3-D canopy was simulated in a physically-based and rigorous way so that various contributions from the radiative process can be accurately quantified. The physical mechanism behind the spectral and angular distributions of canopy SIF was analyzed based on FluorWPS. SIF anisotropy is an intrinsic property of the vegetative surface and it can be significantly influenced by the canopy structure. The performance of the model was evaluated with field measurements and systematically compared with an established RT model of canopy SIF. Especially the detailed comparison with the RT model for four canopy scenes demonstrates that FluorWPS is capable of faithfully reproducing the spectral and angular distributions of SIF, with the coefficient of determination (R-2) and root mean square error (RMSE) being higher than 0.92 and lower than 0.066 W . m(-2). sr(-1) . mu m(-1), respectively, for the red peak, and higher than 0.92 and lower than 0.16 W . m(-2). sr(-1) . mu m(-1) for the far-red peak. The independent nature of the model's canopy realization from the radiative processes is promising for its wide applications for scientific investigations. (C) 2016 Elsevier Inc. All rights reserved.

【 授权许可】

Free