The Leaf Area Index (LAI) is a key indicator for assessing fruit tree growth and productivity, and accurate estimation using hyperspectral technology is essential for monitoring plant health. This study aimed to improve LAI estimation accuracy in apricot, jujube, and walnut trees in Xinjiang, China. Canopy hyperspectral data were processed using fractional-order differentiation (FOD) from 0 to 2.0 orders to extract spectral features. Three feature selection methods—Competitive Adaptive Reweighted Sampling (CARS), Successive Projections Algorithm (SPA), and their combination (CARS-SPA)—were applied to identify sensitive spectral bands. Various band combinations were used to construct three-band indices (TBIs) for optimal LAI estimation. Random forest (RF) models were developed and validated for LAI prediction. The results showed that (1) the reflectance spectra of jujube and walnut trees were similar, while apricot spectra differed. (2) The correlation between fractional-order differential spectra and LAI was highest at orders 1.4 and 1.7, outperforming integer-order spectra. (3) The CARS-SPA selected a smaller set of feature bands in the 1100~2500 nm, reducing collinearity and improving spectral index construction. (4) The RF model using TBI4 demonstrated high R², low RMSE, and an RPD value > 2, indicating optimal prediction accuracy. This approach holds promise for hyperspectral LAI monitoring in fruit trees.

The accurate and timely detection of leaf phosphorus content (LPC) is extremely important for the fertilization management of crop growth and yield. This study aimed to establish an estimating model of LPC in cotton based on hyperspectral data. Under field experimental conditions with different phosphorus treatments, the spectral data and LPC were measured. The spectral characteristics of different cotton cultivars and leaves with varying phosphorus content were analyzed. Optimized spectral indices most correlated to phosphorus were calculated with combinations of arbitrary bands using the Fractional Differential Order (FOD) transform. Then, the random forest-based(RF) estimation model for cotton LPC was established. The research results indicated that (1) the spectral changes of 24 cotton cultivars were basically consistent, and spectral differences between the cultivars became more obvious within the 760–960 nm spectral region; (2) in the visible region, the reflectance of cotton under different phosphorus treatments did not show obvious regularity, while in NIR, the reflectance of cotton increased with the increase in phosphorus content, showing a certain difference in phosphorus; (3) the RF model using a difference spectral index (DSI) had the best performance for LPC estimations in calibration (R² = 0.78) and validation (R² = 0.85), which was superior to the other models based on two spectral indices (the NDSI and RSI). This study provides technical support for the hyperspectral estimation of LPC in cotton.