Quantifying the impact of grasshoppers: an integrated approach using dynamic vegetation indices, red edge shift and...

In conventional remote sensing monitoring, the quantitative estimation of vegetation damage caused by locust feeding has long been challenged by the spatiotemporal resolution limits of satellites, as well as a lack of understanding regarding the physiological characteristics of different locust species. To address this, cage experiments were conducted in a typical steppe region of northern China, with the primary grassland pest in the area, Oedaleus decorus asiaticus, as the study subject.

Field experiments in Xilinhot revealed that vegetation indices were significantly more sensitive to biomass loss at certain stages of locust development than at others. During the fourth instar stage, the ratio vegetation index (RVI) and green normalized difference vegetation index (GNDVI) exhibited high sensitivity to forage loss (R² > 0.79). In the adult stage, the modified soil-adjusted vegetation index (MSAVI) demonstrated greater predictive accuracy for locust feeding intensity (R² > 0.95). The incorporation of natural pasture growth dynamics through the introduction of a vegetation index loss component (ΔVI') effectively decouples environmental background interference, thereby enhancing both the accuracy and robustness of the model. Furthermore, with increasing pest population density, a decline in near-infrared reflectance is observed, accompanied by a shift of the red-edge position toward shorter wavelengths, a phenomenon commonly referred to as 'blue shift.'

Integrating the developmental phenology of locusts with multi-temporal vegetation indices and red-edge position dynamics enables real-time monitoring of grassland locust infestations, quantitative damage assessment, and early warning capabilities. © 2026 Society of Chemical Industry.