Nondestructive determination of leaf chlorophyll content in two flowering cherries using reflectance and absorptance spectra

Leaf chlorophyll quantification is a key technique in tree vigor assessment. Although many studies have been conducted on nondestructive and in-field spectroscopic determination, it is reasonable to develop species-specific chlorophyll indices for accurate determination, because leaf spectra can vary independently of chlorophyll content due to leaf surface and structural differences among species. The present study aimed to develop optimal reflectance and absorptance indices for estimating the leaf chlorophyll content of Cerasus jamasakura (Siebold ex Koidz.) H. Ohba var. jamasakura and Cerasus × yedoensis ‘Somei-yoshino,’ and to examine their performance by comparing them with 46 published chlorophyll indices and SPAD. For 96 and 100 leaf samples, measurements were taken using a spectroradiometer with a leaf-clip attachment and a SPAD-502 chlorophyll meter, and chlorophyll content was determined by extraction with N,N′-dimethylformamide. The optimal leaf chlorophyll indices were then developed systematically by testing eight types of indices. As a result, we confirmed that the optimal chlorophyll indices performed better than any of the published leaf chlorophyll indices or SPAD, giving RMSEs that were approximately twice as good as those for SPAD, and found that the newly proposed index type—a difference and ratio combination type—may be a useful form of chlorophyll content estimation. We also found that absorptance indices achieved equivalent results to reflectance indices despite the hypothesis that absorptance measurement is direct and has more potential. Among the published indices, the reflectance ratio index of Datt [Datt B (1999) Int J Remote Sens 20(14):2741–2759] and the red edge chlorophyll index of Ciganda et al. [Ciganda V, Gitelson A, Schepers J (2009) J Plant Physiol 166:157–167] were effective at estimating the leaf chlorophyll contents of both flowering cherries.