Monitoring the chlorophyll fluorescence parameters in rice under flooding and waterlogging stress based on remote sensing

Flood and waterlog disaster is one of the most serious catastrophes for rice in China. Timely and accurately monitoring waterlogging damage can provide quantitative damage assessment and support for after-flood field management. Chlorophyll fluorescence (CF) is directly related to the waterlogging stress. This paper aims to establish models to monitor the change of chlorophyll fluorescence parameters (FPs) at different growth stages under waterlogging stress based on hyperspectral data. Waterlogging stress was simulated in experimental environment. Back Propagation Neural Network (BPNN) model were proposed by analyzing the relationship between chlorophyll fluorescence parameters (FPs) and spectra absorption feature parameters, which were extracted from continuum removal spectra (550nm-750nm) to represent absorption features. The experimental results indicated that absorption feature parameters and BPNN can improve the estimation accuracy of FPs under flooding and waterlogging stress.