Non-Destructive Assessment of Nutrient Status in ‘Nashi’ Pear Trees Using Optical Methods

Efficient nutrient management is essential for sustainable orchard production; however, conventional laboratory analyses used to assess plant nutritional status are time-consuming and expensive. Optical sensing technologies offer a rapid and non-destructive alternative. This study evaluated the potential of proximal optical sensors and UAV-based multispectral imagery to assess the nutritional status of young potted ‘Nashi’ pear (Pyrus pyrifolia (Burm. f.) Nakai) trees. Three fertilization treatments based on different concentrations of Hoagland solution were applied to 18 one-year-old potted trees. Leaf measurements were collected during the growing season using Dualex, CL-01 chlorophyll meter, and Pocket PEA fluorimeter, while UAV-based multispectral imagery was used to calculate vegetation indices, including NDVI, SR, OSAVI, and MSAVI. Leaf nitrogen (N), phosphorus (P), and potassium (K) concentrations were chemically determined and used as reference values for the regression analyses. Significant (p < 0.05) relationships were observed between leaf N content (N%) and several optical parameters related to leaf pigments, including chlorophyll, flavonols, and the Nitrogen Balance Index (NBI), as well as multispectral indices, although with weak associations (R2 = 0.326–0.488). The strongest individual relationship with N% was shown by NBI (R2 = 0.480). To account for repeated measurements on the same plants, linear mixed-effects models were fitted. These models indicated that NBI showed the strongest association with N% among the proximal optical parameters (β = 0.019, p < 0.001; RMSE = 0.113; MAE = 0.091), followed by flavonols and Dualex chlorophyll. In contrast, optical parameters showed limited sensitivity to P and K. Multispectral indices were not significantly related to K, while only Red and Green reflectance showed weak correlations with P. Overall, optical parameters showed the best associations with N% under the combined nutrient-gradient conditions tested, whereas the assessment of P and K remained limited and should be considered exploratory.