Discriminating C3 and C4 plants from hyperspectral data

This study presents a novel passive-detection method to discriminate between the C3 and C4 plant functional types using the solar-induced ChlF. The solar-induced ChlF radiation was extracted from airborne hyperspectral data acquired by the OMIS II instrument. Our results showed that the fluorescence signal of C4 species was about 2.2 times greater than that of C3 species at the same NDVI level. The spectral separation rate of the solar-induced ChlF was 1.31, but the maximum value of the spectral reflectance and NDVI was only 0.63 at the 750 nm band. A simple decision tree was built to discriminate between the C3 and C4 species based on the difference between their ChlF. An accuracy assessment indicated that the C3 and C4 species had been well classified, with an overall classification accuracy of 92% and a kappa coefficient of 0.84. Although it is quite challenging to classify the C3 and C4 species based on the reflectance signal, our results present a novel method for successfully discriminating between the C3 and C4 species.