Search

Loess environmental parameters reflect regional environmental evolution and provide basis for regional environmental change studies. Loess samples from different soil layers in Zhengzhou were collected, and their hyperspectral data measured. The characteristics of the original spectra, first-order differential spectra, second-order differential spectra, de-enveloping line spectra, and reciprocal logarithm spectra were analysed and correlated with environmental parameters, including mean particle size, standard deviation, Fe2O3, CaO, MgO, and total organic carbon content, based on partial least squares regression. Inversion models were constructed. After transformation, the correlations between the various spectra and the environmental parameters were enhanced. Among them, the first-order and second-order differential spectra showed the greatest correlation with average particle size (correlation coefficient: 0.74). The correlation coefficients of the order differential spectrum with standard deviation, Fe2O3, and CaO were − 0.69, 0.67, and 0.75, respectively; the accuracy grade of the original spectrum inversion model was D. After spectral transformation, it was mostly improved; the best inversion model accuracy grade for standard deviation was C, that of the other parameters was B. Partial transformation is the optimal spectral transformation to establish an inversion model for these environmental parameters. This provides a new soil parameter inversion and monitoring method using hyperspectral remote sensing technology. [ABSTRACT FROM AUTHOR]