Accurate estimation of tree position, diameter at breast height (DBH), and tree height measurements is an important task in forest inventory. Mobile Laser Scanning (MLS) is an important solution. However, the poor global navigation satellite system (GNSS) coverage under the canopy makes the MLS system unable to provide globally-consistent point cloud data, and thus, it cannot accurately estimate the forest attributes. SLAM could be an alternative for solutions dependent on GNSS. In this paper, a mobile phone with RGB-D SLAM was used to estimate tree position, DBH, and tree height in real-time. The main aims of this paper include (1) designing an algorithm to estimate the DBH and position of the tree using the point cloud from the time-of-flight (TOF) camera and camera pose; (2) designing an algorithm to measure tree height using the perspective projection principle of a camera and the camera pose; and (3) showing the measurement results to the observer using augmented reality (AR) technology to allow the observer to intuitively judge the accuracy of the measurement results and re-estimate the measurement results if needed. The device was tested in nine square plots with 12 m sides. The tree position estimations were unbiased and had a root mean square error (RMSE) of 0.12 m in both the x-axis and y-axis directions; the DBH estimations had a 0.33 cm (1.78%) BIAS and a 1.26 cm (6.39%) root mean square error (RMSE); the tree height estimations had a 0.15 m (1.08%) BIAS and a 1.11 m (7.43%) RMSE. The results showed that the mobile phone with RGB-D SLAM is a potential tool for obtaining accurate measurements of tree position, DBH, and tree height.

Global navigation satellite systems (GNSS) can quickly, efficiently, and accurately provide precise coordinates of points, lines, and surface elements, plus complete surveys and determine various boundary lines in forest investigations and management. The system has become a powerful tool for dynamic forest resource investigations and monitoring. GNSS technology plays a unique and important role in estimating timber volume, calculating timber cutting area, and determining the location of virgin forest roads and individual trees in forests. In this study, we quantitatively analyzed the influence of crown size and observation time on the single-point positioning accuracy of GNSS receivers for different forest types. The GNSS located single points for different forest types and crown sizes, enabling the collection of data. The locating time for each tree was more than 10 min. Statistical methods were used to analyze the positioning accuracy of multi-epoch data, and a model was developed to estimate the maximum positioning errors under different forest conditions in a certain positioning time. The results showed that for a continuous positioning time of approximately 10 min, the maximum positioning accuracies in coniferous and broadleaf forests were obtained, which were 12.13 and 15.11 m, respectively. The size of a single canopy had no obvious influence on the single-point positioning error of the GNSS, and canopy density was proven to be closely related to the positioning accuracy of a GNSS. The determination coefficients (R²) in the regression analysis of the general model, coniferous forest model, and broadleaved forest model that were developed in this study were 0.579, 0.701, and 0.544, respectively. These results indicated that the model could effectively predict the maximum positioning error in a certain period of time under different forest types and crown conditions at middle altitudes, which has important guiding significance for forest resource inventories and precise forest management.

Landslides are a common and costly geological hazard, with regular occurrences leading to significant damage and losses. To effectively manage land use and reduce the risk of landslides, it is crucial to conduct susceptibility assessments. To date, many machine-learning methods have been applied to the landslide susceptibility map (LSM). However, as a risk prediction, landslide susceptibility without good interpretability would be a risky approach to apply these methods to real life. This study aimed to assess the LSM in the region of Nayong in Guizhou, China, and conduct a comprehensive assessment and evaluation of landslide susceptibility maps utilizing an explainable artificial intelligence. This study incorporates remote sensing data, field surveys, geographic information system techniques, and interpretable machine-learning techniques to analyze the sensitivity to landslides and to contrast it with other conventional models. As an interpretable machine-learning method, generalized additive models with structured interactions (GAMI-net) could be used to understand how LSM models make decisions. The results showed that the GAMI-net model was valid and had an area under curve (AUC) value of 0.91 on the receiver operating characteristic (ROC) curve, which is better than the values of 0.85 and 0.81 for the random forest and SVM models, respectively. The coal mining, rock desertification, and rainfall greater than 1300 mm were more susceptible to landslides in the study area. Additionally, the pairwise interaction factors, such as rainfall and mining, lithology and rainfall, and rainfall and elevation, also increased the landslide susceptibility. The results showed that interpretable models could accurately predict landslide susceptibility and reveal the causes of landslide occurrence. The GAMI-net-based model exhibited good predictive capability and significantly increased model interpretability to inform landslide management and decision making, which suggests its great potential for application in LSM.

Forests are an important part of the ecological environment, and changes in forests not only affect the ecological environment of the region but are also an important factor causing landslide disasters. In order to correctly evaluate the impact of forest cover on landslide susceptibility, in this paper, we build an evaluation model for the contribution of forests to the landslide susceptibility of different grades based on survey data for forest land change in Bijie City and landslide susceptibility data, and discuss the effects of forest land type, origin, age group, and dominant tree species on landslide susceptibility. We find that forests play a certain role in regulating landslide susceptibility: compared with woodland, the landslide protection ability of shrubland is stronger. Furthermore, natural forests have a greater inhibitory effect on landslides than artificial forests, and compared with young forest, mature forest and over-mature forest, middle-aged forest and near-mature forest have stronger landslide protection abilities. In addition, the dominant tree species in different regions have different impacts on landslides. Coniferous forests such as Chinese fir and Cryptomeria fortunei in Qixingguan and Dafang County have a low ability to prevent landslides. Moreover, the soft broad tree species found in Qianxi County, Zhijin County, Nayong County and Jinsha County are likely to cause landslides and deserve further research attention. Additionally, a greater focus should be placed on the landslide protection of walnut economic forests in Hezhang County and Weining County. Simultaneously, greater attention should be paid to the Cyclobalanopsis glauca tree species in Weining County because the area where this tree species is located is prone to landslides. Aiming at addressing the landslide susceptibility existing in different forests, we propose forest management strategies for the ecological prevention and control of landslides in Bijie City, which can be used as a reference for landslide susceptibility prevention and control.

Incremental vertical ground movements due to coal mining can increase landslide susceptibility greatly in a short time and have thus triggered a large number of geological disasters, especially in the Karst Region, where a lot of steep slopes are on fractured rocks. Therefore, the landslide susceptibility maps (LSM) in Karst Region should be updated regularly. This paper presents an efficient methodology to update and refine LSM by using Persistent Scatterer Interferometry (PSI) data directly. First, an original LSM was produced by using a support vector machine (SVM) algorithm, and the distribution of coal mining was considered a crucial factor to generate the LSM. Then, the Permanent Scatterer Interferometric Synthetic Aperture Radar (PSInSAR) technique was implemented to retrieve displacement time-series. Finally, the landslide displacement map, produced by the PSInSAR analysis, was projected to the direction of the steepest slope and resampled to the same cell in the LSM, to update the original LSM. This methodology is illustrated with the case study of Bijie in the Karst Region of Southwest China, wherein the ascending RADARSAT-2 and descending Sentinel-1 datasets are processed for the period of 2017–2019. The results show that the susceptibility degree increased in 56.41 km² of the study area, and 80 percent of the increased susceptibility degree was caused by coal mining. The comparison between original and refined LSM in two specific areas revealed that the proposed method can produce more-reliable landslide susceptibility maps in areas of intense mining activities in the Karst Region.