Quantitative analysis of chlorophyll in Catalpa bungei leaves based on partial least squares regression and spectral reflectance index

Abstract

The photosynthetic pigment content of forest tree leaves is an important factor affecting photosynthesis in forest trees, and the non-destructive measurement of photosynthetic pigment content is particularly important in current research. The aim of this study was to develop a reflectance index for the determination of photosynthetic pigment content of Catalpa bungei leaves, and in addition to construct a model that can more accurately invert the photosynthetic pigment content. The correlation coefficient method was used to select influential reflectance indices for unitary linear regression model construction; and PLSR was used to accurately describe the photosynthetic pigment content of Catalpa bungei leaves using hyperspectral reflectance and its derived data as input variables. The results showed that the RPD values of Chla and Chl constructed by SR and ND were greater than 2.0, and the RPD value of Chlb was close to 2.0, and the RPD value of Chl/Car was greater than 1.4, and the RPD value of Chla/Chlb was less than 1.4. RLR as a factor to describe the ability of photosynthetic pigment ratio is relatively better than other input variables, Chla concentration and Chlb concentration real inversion to FDRL obviously optimal as input variables, R as the input variables to describe the ability of photosynthetic pigment is relatively better than other input variables. This study constructed an accurate and simplified model for the content of photosynthetic pigments in the leaves of Catalpa bungei, providing technical support for the development of low-cost instruments and the intrinsic correlation between photosynthetic pigments and reflectance.