Use of electronic nose technology for identifying rice infestation by Nilaparvata lugens

Abstract

Plant-emitted volatiles can change after herbivore attack. Monitoring the change in volatile profiles can offer a non-destructive method for determining plant health. An electronic nose (E-nose) equipped with a headspace sampling unit was used to discriminate between volatile profiles emitted by uninfested rice plants and those emitted by rice plants exposed to different numbers of Nilaparvata lugens adults. Principal component analysis (PCA) and linear discriminant analysis (LDA) were used to investigate whether the E-nose was able to distinguish among the different pest treatments. The results indicate that it is possible to separate differently treated rice plants using E-nose signals. The stepwise discriminant analysis (SDA) and a 3-layer back-propagation neural network (BPNN) were developed for pattern recognition models. Calculations show that the discrimination rates were over 92.5% for the training data set and 70% for the testing set using SDA. The correlation coefficient between predicted and real numbers of the pest was found to be over 0.78 using BPNN. Moreover, gas chromatography–mass spectrometry (GC–MS) analysis confirmed the E-nose results. These studies demonstrate that the E-nose technology has clear potential for use as an effective insect monitoring method.