Using high-resolution remote sensing data to identify infected trees is an important method for controlling pine wilt disease (PWD). Currently, single-date image classification methods are widely used for PWD detection in pure stands of pine. However, they often yield false detections caused by deciduous trees, brown herbaceous, and sparsely vegetated regions in complex landscapes, resulting in low user accuracies. Due to the limitations on the bands of the high-resolution imagery, it is difficult to distinguish wilted pine trees from such easily confused objects when only using the optical spectral characteristics. This paper proposes a spatiotemporal change detection method to reduce false detections in tree-scale PWD monitoring under a complex landscape. The framework consisted of three parts, which represent the capture of spectral, temporal, and spatial features: (1) the Normalized Green–Red Difference Index (NGRDI) was calculated as a descriptor of canopy greenness; (2) two NGRDI images with similar dates in adjacent years were contrasted to obtain a bitemporal change index that represents the temporal behaviors of typical cover types; and (3) a spatial enhancement was performed on the change index using a convolution kernel matching the spatial patterns of PWD. Finally, a set of criteria based on the above features were established to extract the wilted pine trees. The results showed that the proposed method effectively distinguishes wilted pine trees from other easily confused objects. Compared with single-date image classification, the proposed method significantly improved user’s accuracy (81.2% vs. 67.7%) while maintaining the same level of producer’s accuracy (84.7% vs. 82.6%).

Pine wilt disease (PWD) has caused a huge damage to pine forests. PWD is mainly transmitted by jumping diffusion, affected by insect vectors and human activities. Since the results of climate change, pine wood nematode (PWN—Bursaphelenchus xylophilus) has begun invading the temperate zones and higher elevation area. In this situation, predicting the distribution of PWD is an important part of the prevention and control of the epidemic situation. The research established the Maxent model to conduct a multi-angle, fine-scale prediction on the risk distribution of PWD. We adjusted two parameters, regularization multiplier (RM) and feature combination (FC), to optimize the model. Influence factors were selected and divided into natural, landscape, and human variables, according to the physical characteristics and spread rules of PWD. The middle-suitability regions and high-suitability regions are distributed in a Y-shape, and divided the study area into three parts. The high-suitability areas are concentrated in the region with high temperature, low elevation, and intensive precipitation. Among the selected variables, natural factors still play the most important role in the distribution of the disease, and human factors and landscape factors are also worked well. The permutation importance of factors is different due to differences in climate and other conditions in different regions. The multi-angle, fine-scale model can help provide useful information for effective control and tactical management of PWD.

Grasshoppers can greatly interfere with agriculture and husbandry, and they will breed and grow rapidly in suitable habitats. Therefore, it is necessary to extract the distribution of the grasshopper potential habitat (GPH), analyze the spatial-temporal characteristics of the GPH, and detect the different effects of key environmental factors in the meadow and typical steppe. To achieve the goal, this study took the two steppe types of Xilingol (the Inner Mongolia Autonomous Region of China) as the research object and coupled them with the MaxEnt and multisource remote sensing data to establish a model. First, the environmental factors, including meteorological, vegetation, topographic, and soil factors, that affect the developmental stages of grasshoppers were obtained. Secondly, the GPH associated with meadow and typical steppes from 2018 to 2022 were extracted based on the MaxEnt model. Then, the spatial-temporal characteristics of the GPHs were analyzed. Finally, the effects of the habitat factors in two steppe types were explored. The results demonstrated that the most suitable and moderately suitable areas were distributed mainly in the southern part of the meadow steppe and the eastern and southern parts of the typical steppe. Additionally, most areas in the town of Gaorihan, Honggeergaole, Jirengaole, as well as the border of Wulanhalage and Haoretugaole became more suitable for grasshoppers from 2018 to 2022. This paper also found that the soil temperature in the egg stage, the vegetation type, the soil type, and the precipitation amount in the nymph stage were significant factors both in the meadow and typical steppes. The slope and precipitation in the egg stage played more important roles in the typical steppe, whereas the aspect had a greater contribution to the meadow steppe. These findings can provide a methodical guide for grasshopper control and management and for further ensuring the security of agriculture and husbandry.

The fall armyworm (Spodoptera frugiperda) (J. E. Smith) is a widespread, polyphagous, and highly destructive agricultural pest. Global climate change may facilitate its spread to new suitable areas, thereby increasing threats to host plants. Consequently, predicting the potential suitable distribution for the fall armyworm and its host plants under current and future climate scenarios is crucial for assessing its outbreak risks and formulating control strategies. This study, based on remote sensing assimilation data and plant protection survey data, utilized machine learning methods (RF, CatBoost, XGBoost, LightGBM) to construct potential distribution prediction models for the fall armyworm and its 120 host plants. Hyperparameter methods and stacking ensemble method (SEL) were introduced to optimize the models. The results showed that SEL demonstrated optimal performance in predicting the suitable distribution for the fall armyworm, with an AUC of 0.971 ± 0.012 and a TSS of 0.824 ± 0.047. Additionally, LightGBM and SEL showed optimal performance in predicting the suitable distribution for 47 and 30 host plants, respectively. Overlay analysis suggests that the overlap areas and interaction links between the suitable areas for the fall armyworm and its host plants will generally increase in the future, with the most significant rise under the RCP8.5 climate scenario, indicating that the threat to host plants will further intensify due to climate change. The findings of this study provide data support for planning and implementing global and intercontinental long-term pest management measures aimed at mitigating the impact of the fall armyworm on global food production.

Powdery mildew is one of the most significant rubber tree diseases, with a substantial impact on the yield of natural rubber. This study aims to establish a detection approach that coupled continuous wavelet transform (CWT) and machine learning for the accurate assessment of powdery mildew severity in rubber trees. In this study, hyperspectral reflectance data (350–2500 nm) of healthy and powdery mildew-infected leaves were measured with a spectroradiometer in a laboratory. Subsequently, three types of wavelet features (WFs) were extracted using CWT. They were as follows: WFs dimensionally reduced by the principal component analysis (PCA) of significant wavelet energy coefficients (PCA-WFs); WFs extracted from the top 1% of the determination coefficient between wavelet energy coefficients and the powdery mildew disease class (1%R²-WFs); and all WFs at a single decomposition scale (SS-WFs). To assess the detection capability of the WFs, the three types of WFs were input into the random forest (RF), support vector machine (SVM), and back propagation neural network (BPNN), respectively. As a control, 13 optimal traditional spectral features (SFs) were extracted and combined with the same classification methods. The results revealed that the WF-based models all performed well and outperformed those based on SFs. The models constructed based on PCA-WFs had a higher accuracy and more stable performance than other models. The model combined PCA-WFs with RF exhibited the optimal performance among all models, with an overall accuracy (OA) of 92.0% and a kappa coefficient of 0.90. This study demonstrates the feasibility of combining CWT with machine learning in rubber tree powdery mildew detection.

Powdery mildew significantly impacts the yield of natural rubber by being one of the predominant diseases that affect rubber trees. Accurate, non-destructive recognition of powdery mildew in the early stage is essential for the cultivation management of rubber trees. The objective of this study is to establish a technique for the early detection of powdery mildew in rubber trees by combining spectral and physicochemical parameter features. At three field experiment sites and in the laboratory, a spectroradiometer and a hand-held optical leaf-clip meter were utilized, respectively, to measure the hyperspectral reflectance data (350–2500 nm) and physicochemical parameter data of both healthy and early-stage powdery-mildew-infected leaves. Initially, vegetation indices were extracted from hyperspectral reflectance data, and wavelet energy coefficients were obtained through continuous wavelet transform (CWT). Subsequently, significant vegetation indices (VIs) were selected using the ReliefF algorithm, and the optimal wavelengths (OWs) were chosen via competitive adaptive reweighted sampling. Principal component analysis was used for the dimensionality reduction of significant wavelet energy coefficients, resulting in wavelet features (WFs). To evaluate the detection capability of the aforementioned features, the three spectral features extracted above, along with their combinations with physicochemical parameter features (PFs) (VIs + PFs, OWs + PFs, WFs + PFs), were used to construct six classes of features. In turn, these features were input into support vector machine (SVM), random forest (RF), and logistic regression (LR), respectively, to build early detection models for powdery mildew in rubber trees. The results revealed that models based on WFs perform well, markedly outperforming those constructed using VIs and OWs as inputs. Moreover, models incorporating combined features surpass those relying on single features, with an overall accuracy (OA) improvement of over 1.9% and an increase in F1-Score of over 0.012. The model that combines WFs and PFs shows superior performance over all the other models, achieving OAs of 94.3%, 90.6%, and 93.4%, and F1-Scores of 0.952, 0.917, and 0.941 on SVM, RF, and LR, respectively. Compared to using WFs alone, the OAs improved by 1.9%, 2.8%, and 1.9%, and the F1-Scores increased by 0.017, 0.017, and 0.016, respectively. This study showcases the viability of early detection of powdery mildew in rubber trees.