Hyperspectral remote sensing to quantify the flowering phenology of winter wheat

Abstract

Flowering phenology, as a sensitive bio-indicator to climate change, has been attracting increasing concern. However, less information is available so far for quantifying flowering phenological stages by use of hyperspectral remote sensing in winter wheat. The establishment of a vegetation index, which is sensitive to the changes of flowering rate, is important to allow the monitoring of flowering phenology through hyperspectral remote sensing. A field experiment was conducted at the Agro-meteorological Station, Nanjing University of Information Science and Technology, Jiangsu Province, China, with two nighttime warming levels, i.e., nighttime warming and ambient temperature (control), and two sowing levels, i.e., normal sowing and late sowing. The field canopy reflectance spectrum was synchronously determined with the flowering rate within 0.25 m2 area at the flowering stage of winter wheat. This study developed a new index for estimating the flowering rate that was compared with the published indices for estimating canopy structure parameter. The newly developed two bands-index (R446 − R472)/(R446 + R472), generated coefficients of determination, root mean square error, and residual prediction deviation values between the measured and predicted values of 0.75, 18.69%, and 1.66, respectively. The study suggests that hyperspectral remote sensing can be used to quantify flowering phenology in winter wheat.