Spectral characterization and severity assessment of rice brown planthopper damage using multivariate models

Brown planthopper (BPH) is a serious rice pest that threatens global food security by causing yield losses of up to 80%. Conventional methods for assessing BPH infestation are labour-intensive and lack real-time precision. This study evaluates hyperspectral remote sensing as a rapid, non-invasive approach for quantifying BPH population severity in three rice varieties: Pusa Basmati-1509, Pusa Basmati-1121, and TN-1. Leaf-level spectral measurements (350–2500 nm) acquired using a portable spectroradiometer effectively differentiated BPH population severity levels. Among 28 spectral indices evaluated, Structural Insensitive Pigment Index (SIPI), Pigment Specific Normalized Difference Index (PSND) for chlorophyll b, Pigment Specific Simple Ratio (PSSR a) for chlorophyll a, and (PSSR b) for chlorophyll b, showed high sensitivity to BPH infestation. Multivariate Regression models, including Partial Least Squares Regression (PLSR), Support Vector Machine (SVM), and Random Forest (RF), were developed for severity prediction. Among the tested models, RF achieved the highest accuracy for vegetation indices-based estimation (R2 = 0.99), while PLSR showed strong relationships between hyperspectral data and BPH population severity (R2 = 0.62) and key biochemical parameters, including chlorophyll (R2 = 0.84), carotenoids (R2 = 0.77), and protein (R2 = 0.84). In contrast, flavonoids exhibited weak predictability (R2 = 0.34). Field validation confirmed model robustness, with vegetation index-based predictions achieving R2 values ranging from 0.72 to 0.86. Overall, the results demonstrate the potential of hyperspectral sensing combined with machine learning for early, non-destructive detection and monitoring of BPH stress, supporting precision pest management in rice.