As one of the most devastating disasters to pine forests, pine wilt disease (PWD) has caused tremendous ecological and economic losses in China. An effective way to prevent large-scale PWD outbreaks is to detect and remove the damaged pine trees at the early stage of PWD infection. However, early infected pine trees do not show obvious changes in morphology or color in the visible wavelength range, making early detection of PWD tricky. Unmanned aerial vehicle (UAV)-based hyperspectral imagery (HI) has great potential for early detection of PWD. However, the commonly used methods, such as the two-dimensional convolutional neural network (2D-CNN), fail to simultaneously extract and fully utilize the spatial and spectral information, whereas the three-dimensional convolutional neural network (3D-CNN) is able to collect this information from raw hyperspectral data. In this paper, we applied the residual block to 3D-CNN and constructed a 3D-Res CNN model, the performance of which was then compared with that of 3D-CNN, 2D-CNN, and 2D-Res CNN in identifying PWD-infected pine trees from the hyperspectral images. The 3D-Res CNN model outperformed the other models, achieving an overall accuracy (OA) of 88.11% and an accuracy of 72.86% for detecting early infected pine trees (EIPs). Using only 20% of the training samples, the OA and EIP accuracy of 3D-Res CNN can still achieve 81.06% and 51.97%, which is superior to the state-of-the-art method in the early detection of PWD based on hyperspectral images. Collectively, 3D-Res CNN was more accurate and effective in early detection of PWD. In conclusion, 3D-Res CNN is proposed for early detection of PWD in this paper, making the prediction and control of PWD more accurate and effective. This model can also be applied to detect pine trees damaged by other diseases or insect pests in the forest.

Pine wilt disease (PWD) has caused huge economic and environmental losses since it invaded China. Although early monitoring is an effective way to control this hazard, the monitoring window for the early stage is hard to identify, and varies in different hosts and environments. We used UAV-based multispectral images of Pinus thunbergii forest in East China to identify the change in the number of infected trees in each month of the growing season. We built classification models to detect different PWD infection stages by testing three machine learning algorithms—random forest, support vector machine, and linear discriminant analysis—and identified the best monitoring period for each infection stage (namely, green attack, early, middle, and late). From the obtained results, the early monitoring window period was determined to be in late July, whereas the monitoring window for middle and late PWD stages ranged from mid-August to early September. We also identified four important vegetation indices to monitor each infection stage. In conclusion, this study demonstrated the effectiveness of using machine learning algorithms to analyze multitemporal multispectral data to establish a window for early monitoring of pine wilt disease infestation. The results could provide a reference for future research and guidance for the control of pine wilt disease.

Tomicus yunnanensis and Tomicus minor have caused serious shoot damage in Yunnan pine forests in the Yunnan Province of China. However, very few remote sensing studies have estimated the shoot damage ratio (SDR). Thus, we used multi-date Landsat satellite imagery to quantify SDRs and assess the possibility of using spectral indices to determine the beetle outbreak time and spread direction. A new threshold-based classification method was proposed to identify damage levels (i.e., healthy, slightly to moderately infested, and severely infested forests) using time series of moisture stress index (MSI). Permanent plots and temporal field inspection data were both used as references for training and evaluation. Results show that a single threshold value can produce a total classification accuracy of 86.38% (Kappa = 0.80). Furthermore, time series maps detailing damage level were reconstructed from 2004 to 2016. The shoot beetle outbreak year was estimated to be 2013. Another interesting finding is the movement path of the geometric center of severe damage, which is highly consistent with the wind direction. We conclude that the time series of shoot damage level maps can be produced by using continuous MSI images. This method is very useful to foresters for determining the outbreak time and spread direction.

Beijing’s One Million Mu Plain Afforestation Project involves planting large areas with the exotic North American tree species Fraxinus pennsylvanica Marsh (ash). As an exotic tree species, ash is very vulnerable to infestations by the emerald ash borer (EAB), a native Chinese wood borer pest. In the early stage of an EAB infestation, attacked trees show no obvious sign. Once the stand has reached the late damage stage, death occurs rapidly. Therefore, there is a need for efficient early detection methods of EAB stress over large areas. The combination of unmanned aerial vehicle (UAV)-based hyperspectral imaging (HI) with light detection and ranging (LiDAR) is a promising practical approach for monitoring insect disturbance. In this study, we identified the most useful narrow-band spectral HI data and 3D LiDAR data for the early detection of EAB stress in ash. UAV-HI data of different infested stages (healthy, light, moderate and severe) of EAB in the 400–1000 nm range were collected from ash canopies and were processed by Partial Least Squares–Variable Importance in Projection (PLS-VIP) to identify the maximally sensitive bands. Band R678 nm had the highest PLS-VIP scores and the most robust classification ability. We combined this band with band R776 nm to develop an innovative normalized difference vegetation index (NDVI_(776,678)) to estimate EAB stress. LiDAR data were used to segment individual trees and supplement the HI data. The new NDVI_(776,678) identified different stages of EAB stress, with a producer’s accuracy of 90% for healthy trees, 76.25% for light infestation, 58.33% for moderate infestation, and 100% for severe infestation, with an overall accuracy of 82.90% when combined with UAV-HI and LiDAR.

Yunnan pine shoot beetles (PSB), Tomicus yunnanensis and Tomicus minor have spread through southwestern China in the last five years, leading to millions of hectares of forest being damaged. Thus, there is an urgent need to develop an effective approach for accurate early warning and damage assessment of PSB outbreaks. Remote sensing is one of the most efficient methods for this purpose. Despite many studies existing on the mountain pine beetle (MPB), very little work has been undertaken on assessing PSB stress using remote sensing. The objective of this paper was to develop a spectral linear mixing model aided by radiative transfer (RT) and a new Yellow Index (YI) to simulate the reflectance of heterogeneous canopies containing damaged needles and quantitatively inverse their PSB stress. The YI, the fraction of dead needles, is a physically-explicit stress indicator that represents the plot shoots damage ratio (plot SDR). The major steps of this methods include: (1) LIBERTY2 was developed to simulate the reflectance of damaged needles using YI to linearly mix the green needle spectra with the dead needle spectra; (2) LIBERTY2 was coupled with the INFORM model to scale the needle spectra to the canopy scale; and (3) a look-up table (LUT) was created against Sentinel 2 (S2) imagery and inversed leaf chlorophyll content (LCC), green leaf area index (LAI) and plot SDR. The results show that (1) LIBERTY2 effectively simulated the reflectance spectral values on infested needles (mean relative error (MRE) = 1.4–18%), and the YI can indicate the degrees of needles damage; (2) the coupled LIBERTY2-INFORM model is suitable to estimate LAI (R² = 0.73, RMSE = 0.17 m m⁻², NRMSE = 11.41% and the index of agreement (IOA) = 0.92) and LCC (R² = 0.49, RMSE = 56.24 mg m⁻², NRMSE = 25.22% and IOA = 0.72), and is better than the original LIBERTY model (LAI: R² = 0.38, RMSE = 0.43 m m⁻², NRMSE = 28.85% and IOA = 0.68; LCC: R² = 0.34, RMSE = 76.44 mg m⁻², NRMSE = 34.23% and IOA = 0.57); and (3) the inversed YI is positively correlated with the measured plot SDR (R² = 0.40, RMSE = 0.15). We conclude that the LIBERTY2 model improved the reflectance simulation accuracy of both the needles and canopies, making it suitable for assessing PSB stress. The YI has the potential to assess PSB damage.

The red turpentine beetle (Dendroctonus valens LeConte) has caused severe ecological and economic losses since its invasion into China. It gradually spreads northeast, resulting in many Chinese pine (Pinus tabuliformis Carr.) deaths. Early detection of D. valens infestation (i.e., at the green attack stage) is the basis of control measures to prevent its outbreak and spread. This study examined the changes in spectral reflectance after initial attacking of D. valens. We also explored the possibility of detecting early D. valens infestation based on spectral vegetation indices and machine learning algorithms. The spectral reflectance of infested trees was significantly different from healthy trees (p < 0.05), and there was an obvious decrease in the near-infrared region (760–1386 nm; p < 0.01). Spectral vegetation indices were input into three machine learning classifiers; the classification accuracy was 72.5–80%, while the sensitivity was 65–85%. Several spectral vegetation indices (DID, CUR, TBSI, DDn₂, D₇₃₅, SR₁, NSMI, R_NIR•CRI₅₅₀ and RVSI) were sensitive indicators for the early detection of D. valens damage. Our results demonstrate that remote sensing technology could be successfully applied to early detect D. valens infestation and clarify the sensitive spectral regions and vegetation indices, which has important implications for early detection based on unmanned airborne vehicle and satellite data.