Inconsistent Diurnal Patterns of Far-Red Solar-Induced Chlorophyll Fluorescence Retrieved with Different Algorithms from Tower-Based Observations

Abstract

Tower-based solar-induced chlorophyll fluorescence (SIF) measurements have yielded crucial datasets for investigating the diurnal patterns of SIF and its relationship with vegetation photosynthesis. This study assessed the performance of 3 distinct SIF retrieval algorithms, including band shape fitting (BSF), 3-band Fraunhofer line discrimination (3FLD), and a data-driven approach based on singular vector decomposition (SVD), for retrieving far-red SIF diurnal patterns from tower-based observations at the 2 flux sites in China. This study analyzed diurnal patterns of SIF and SIF yield, as well as correlations between SIF, near-infrared radiance reflected by vegetation (NIRvR), and gross primary productivity (GPP) at diurnal and seasonal scales. More pronounced inconsistencies in retrieved SIF by different algorithms at noon compared with the morning and afternoon were observed. Similarly, correlations between the SIF and NIRvR or GPP are weaker during midday. This study underscores the need to consider the reliability of SIF data when investigating diurnal patterns, and the necessity for developments in tower-based SIF retrieval algorithms.