Behavioral responses of Platycladus orientalis plant volatiles to Phloeosinus aubei by GC-MS and HS-GC-IMS for discrimination of different invasive severity

In recent years, the invasive cypress bark beetle ( Phloeosinus aubei ) has caused extensive damage to Platycladus orientalis plants in China, but its infestation is hard to monitor in the early stages. In this study, gas chromatography-mass spectrometry (GC-MS) was initially employed to investigate the volatile organic compound (VOC) emissions of P. aubei- infested P. orientalis saplings. The emissions of total sesquiterpenes were dominating (84−86% of total VOCs) and increased by 3.09-fold in P. aubei- damaged P. orientalis samples compared to undamaged samples, and the monoterpenes, aromatic compounds, and ketone emissions also had varying degrees of increase between 1.39-fold and 5.65-fold. Based on this variation, gas chromatography-ion mobility spectrometry (GC-IMS) was applied, as an untargeted analytical approach, to discriminate P. orientalis samples with different invasive severity. Two different features derived from GC-IMS data were adopted as the input information for classification and prediction models. Results showed that grid search support vector machine (GS-SVM) combined with multilinear principal component analysis (MPCA) based on spectral fingerprint achieved the best classification performances (> 88.98%), and partial least squares discriminant analysis (PLSR) method can accurately predict the pest numbers ( R 2 > 0.9423 and RMSE < 0.9827). In a word, the VOC profiling-based approach had the potential for evaluating P. aubei invasive severity and pest management.