Revised photochemical reflectance index (PRI) for predicting light use efficiency of wheat in a growth cycle: validation and comparison
Source: International Journal of Remote Sensing, Volume 31, Number 11, 2010 , pp. 2911-2924(14)
Publisher: Taylor and Francis Ltd
Abstract:The photochemical reflectance index (PRI) was developed to trace the changes in light use efficiency (LUE) as the two contributing reflectances at 531 nm and 570 nm are closely related to the xanthophyll pigment cycle. In this paper, two revised indices of PRI (PRIR1 and PRIR2) are derived for a better prediction of LUE during the growth cycle of wheat. The signal of chlorophyll content (reflectance at 550 nm) to PRI is incorporated so that the revised indices can be used to estimate LUE values at low chlorophyll concentration. A validation was conducted using ground data (reflectance and LUE data) during a growth cycle of wheat in 2007 (17 April, 28 April, 16, 29 May). The results demonstrate that PRI cannot be used as an index for LUE estimation during the growth cycle of wheat as the relationship between PRI and LUE significantly weakened (R2 = 0.20) on 29 May when the leaves lost chlorophyll concentration in the senescent period. PRIR1 and PRIR2 are more robust than PRI for LUE estimationm, not only with a relatively stable precision (R2 = 0.62, 0.76, 0.62, 0.57 for PRIR1 and R2 = 0.62, 0.76, 0.63, 0.59 for PRIR2) but also with better linearity with LUE (standard error of regression equation between LUE and index is 0.00187, 0.00127, 0.00116, 0.00103 for PRIR1 and 0.00186, 0.00117, 0.00114, 0.00102 for PRIR2). The result of the comparison analysis indicates that the revised indices (PRIR1 and PRIR2) are more sensitive than PRI to low chlorophyll content and low leaf area index, which means they are more appropriate for LUE interpretation in these situations. Sensitivity of Sun-sensor geometry to all indices implies that all indices exhibit large variations with changes in solar zenith angle and view zenith angle. As solar zenith angle increases, all indices display different sensitivity patterns before and after hotspot positions. All indices vary greatly as the view zenith angle increases. An acceptable precision of all indices can be acquired within a departure of 10° from the nadir view.
Document Type: Research Article
Affiliations: 1: The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing, China,Graduate University of Chinese Academy of Science, Beijing, China 2: The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing, China 3: The State Key Laboratory of Remote Sensing Science, Institute of Remote Sensing Applications, Chinese Academy of Sciences, Beijing, China,Information Technology in Agriculture, National Engineering Research Center, Beijing, China
Publication date: 2010