Search Results (33)

Benggang (Benggang), a typical landform characterized by severe erosion and a geohazard in the red-soil hilly regions of southern China, is characterized by a fragmented texture, irregular boundaries, and high similarity to background objects such as bare soil and roads, which poses a dual challenge of “multiscale variability + strong noise” for automated identification at regional scales. To address insufficient information from a single modality and the limited representation of cross-scale features, this study proposes a dual-stream feature-fusion network (DF-Net) for multisource data consisting of a digital orthophoto map (DOM) and a digital elevation model (DEM). The method adopts ResNeSt50d as the backbone of the two branches: on the DOM side, a Canny-edge channel is stacked to enhance high-frequency boundary information; on the DEM side, derived terrain factors, including slope, aspect, curvature, and hillshade, are introduced to provide morphological constraints. In the cross-modal fusion stage, a multiscale sparse attention fusion module is designed, which acquires contextual information via multiwindow average pooling and suppresses noise interference through top-K sparsification. In the decision stage, a multibranch ensemble is employed to improve classification stability. Taking Anxi County, Fujian Province, as the study area, a coregistered dataset of GF-2 (1 m) DOM and ALOS (12.5 m) DEMs is constructed, and a zonal partitioning strategy is adopted to evaluate the model’s generalization ability. The experimental results show that DF-Net achieves 97.44% accuracy, 85.71% recall, and an 82.98% F1 score in the independent test zone, outperforming multiple mainstream CNN/transformer classification models. This study indicates that the strategy of “multimodal feature enhancement + sparse attention fusion” tailored to Benggang erosional landforms can significantly improve recognition performance under complex backgrounds, providing technical support for rapid Benggang surveys and governance-effectiveness assessments. Full article

Forest surveying and inspection face significant challenges due to unstructured environments, variable terrain conditions, and the high costs of manual data collection. Although mobile robotics and artificial intelligence offer promising solutions, reliable autonomous navigation in forest, terrain-aware path planning, and tree parameter estimation remain open challenges. In this paper, we present the results of the AI4FOREST project, which addresses these issues through three main contributions. First, we develop an autonomous mobile robot, integrating SLAM-based navigation, 3D point cloud reconstruction, and a vision-based deep learning architecture to enable tree detection and diameter estimation. This system demonstrates the feasibility of generating a digital twin of forest while operating autonomously. Second, to overcome the limitations of classical navigation approaches in heterogeneous natural terrains, we introduce a machine learning-based surrogate model of wheel–soil interaction, trained on a large synthetic dataset derived from classical terramechanics. Compared to purely geometric planners, the proposed model enables realistic dynamics simulation and improves navigation robustness by accounting for terrain–vehicle interactions. Finally, we investigate the impact of point cloud density on the accuracy of forest parameter estimation, identifying the minimum sampling requirements needed to extract tree diameters and heights. This analysis provides support to balance sensor performance, robot speed, and operational costs. Overall, the AI4FOREST project advances the state of the art in autonomous forest monitoring by jointly addressing SLAM-based mapping, terrain-aware navigation, and tree parameter estimation. Full article

Machine learning-based digital soil mapping often struggles with spatial heterogeneity and long-range dependencies. To address these limitations, this study proposes Multi-Attention Convolutional Neural Networks (MACNN). This deep learning algorithm integrates multiple attention mechanisms to improve mapping accuracy. First, environmental covariates are determined from the soil-landscape model. These are then fed as structured input to the Convolutional Neural Network. Next, by incorporating Transformer self-attention and multi-head attention mechanisms, this study effectively models the long-range dependencies between soil types and features. Concurrently, the Convolutional Block Attention Module (CBAM) is introduced. CBAM features both channel and spatial dual attention, enabling adaptive weighting of crucial feature channels and spatial locations. This significantly enhances the algorithm’s sensitivity to discriminative information. To validate its effectiveness, the proposed MACNN algorithm was used for soil type mapping in Heilongjiang Province. Compared to Random Forest, Decision Tree, and One-Dimensional Convolutional Neural Network algorithms, MACNN demonstrated superior classification performance. It achieved an overall classification accuracy of 81.27%. An ablation study was conducted to investigate the importance of individual modules within the proposed algorithm. The findings indicate that progressively integrating Transformer and CBAM modules into the 1D-CNN baseline significantly enhances algorithm performance through synergistic gains. Therefore, this integrated algorithm offers a feasible solution to improve digital soil mapping accuracy, providing significant reference value for future research and applications. Full article

Floods are among the most destructive natural disasters and can devastate human life, infrastructure, and mobility in urban areas. It is necessary to develop a simulation model suitable for disaster management to prepare for flooding and facilitate rapid response interventions. The advantage of a three-dimensional (3D) geographic information system (GIS) is that it allows researchers to perform more successful spatial analyses than traditional two-dimensional (2D) systems. In this study, real-time 3D flood simulations were created for the Bozkurt district of Kastamonu, Türkiye, integrating GIS and game engine technologies. Land use land cover (LU/LC) map, digital elevation model (DEM), soil properties and climate data of the study region constitute the input data for the hydrological model. DEM and building footprints are also used to create 3D models of the buildings in the region. Through the Soil and Water Assessment Tool (SWAT) analysis, a hydrological model that included environmental factors such as precipitation, runoff, and soil erosion was created. The average flow rate for the same period, obtained from flow monitoring stations in the Bozkurt district, was 4.64 m³/s, while the flow rate obtained with the SWAT+ model was 4.12 m³/s. Using the flow parameters obtained with SWAT, 3D flood models were developed on Unreal Engine (UE). The flood simulation created with UE and the flood disaster experienced in 2021 in the region were compared on an area basis. The obtained simulation accuracy was 88%. Full article

To support bearing capacity estimates, this study develops and tests a geoprocessing workflow for predicting soil properties using Empirical Bayesian Kriging 3D and a classification function. The model covers a 183 m × 185 m × 24 m site in Astana (Kazakhstan), based on 16 boreholes (15–24 m deep) and 77 samples. Eight geotechnical properties were mapped in 3D voxel models (812,520 voxels at 1 m × 1 m × 1 m resolution): cohesion (c), friction angle (φ), deformation modulus (E), plasticity index (PI), liquidity index (LI), porosity (e), particle size (PS), and particle size distribution (PSD). Stratification patterns were revealed with ~35% variability. Maximum φ (34.9°), E (36.6 MPa), and PS (1.29 mm) occurred at 8–16 m; c (33.1 kPa) and PSD peaked below 16 m, while PI and e were elevated in the upper and lower strata. Strong correlations emerged in pairs φ-E-PS (0.91) and PI-e (0.95). Classification identified 10 soil types, including one absent in borehole data, indicating the workflow’s capacity to detect hidden lithologies. Predicted fractions of loams (51.99%), sandy loams (22.24%), and sands (25.77%) matched borehole data (52%, 26%, 22%). Adjacency analysis of 2,394,873 voxel pairs showed homogeneous zones in gravel–sandy soils (28%) and stiff loams (21.75%). The workflow accounts for lateral and vertical heterogeneity, reduces subjectivity, and is recommended for digital subsurface 3D mapping and construction design optimization. Full article

This research proposes an artificial intelligence (AI)-powered digital twin framework for highway slope stability risk monitoring and prediction. For highway slope stability, a digital twin replicates the geological and structural conditions of highway slopes while continuously integrating real-time monitoring data to refine and enhance slope modeling. The framework employs instance segmentation and a random forest model to identify embankments and slopes with high landslide susceptibility scores. Additionally, artificial neural network (ANN) models are trained on historical drilling data to predict 3D subsurface soil type point clouds and groundwater depth maps. The USCS soil classification-based machine learning model achieved an accuracy score of 0.8, calculated by dividing the number of correct soil class predictions by the total number of predictions. The groundwater depth regression model achieved an RMSE of 2.32. These predicted values are integrated as input parameters for seepage and slope stability analyses, ultimately calculating the factor of safety (FoS) under predicted rainfall infiltration scenarios. The proposed methodology automates the identification of embankments and slopes using sub-meter resolution Light Detection and Ranging (LiDAR)-derived digital elevation models (DEMs) and generates critical soil properties and pore water pressure data for slope stability analysis. This enables the provision of early warnings for potential slope failures, facilitating timely interventions and risk mitigation. Full article

Large-scale mapping of soil resources can be crucial and indispensable for several of the managerial applications and policy implications. With machine learning models being the most utilized modeling technique for digital soil mapping (DSM), the implementation of model-based deep learning methods for spatial soil predictions is still under scrutiny. In this study, soil continuous (pH and OC) and categorical variables (order and suborder) were predicted using deep learning–multi layer perceptron (DL-MLP) and one-dimensional convolutional neural networks (1D-CNN) for the entire state of Tamil Nadu, India. For training the deep learning models, 27,098 profile observations (0–30 cm) were extracted from the generated soil database, considering soil series as the distinctive stratum. A total of 43 SCORPAN-based environmental covariates were considered, of which 37 covariates were retained after the recursive feature elimination (RFE) process. The validation and test results obtained for each of the soil attributes for both the algorithms were most comparable with the DL-MLP algorithm depicting the attributes’ most intricate spatial organization details, compared to the 1D-CNN model. Irrespective of the algorithms and datasets, the R² and RMSE values of the pH attribute ranged from 0.15 to 0.30 and 0.97 to 1.15, respectively. Similarly, the R² and RMSE of the OC attribute ranged from 0.20 to 0.39 and 0.38 to 0.42, respectively. Further, the overall accuracy (OA) of the order and suborder classification ranged from 39% to 67% and 35% to 64%, respectively. The explicit quantification of the covariate importance derived from the permutation feature importance implied that both the models tried to incorporate the covariate importance with respect to the genesis of the soil attribute under study. Such approaches of the deep learning models integrating soil–environmental relationships under limited parameterization and computing costs can serve as a baseline study, emphasizing opportunities in increasing the transferability and generalizability of the model while accounting for the associated environmental dependencies. Full article

We present a data-driven, in situ proximal multi-sensor digital soil mapping approach to develop digital twins for multiple agricultural fields. A novel Digital Soil Core^TM (DSC) Probe was engineered that contains seven sensors, each of a distinct modality, including sleeve friction, tip force, dielectric permittivity, electrical resistivity, soil imagery, acoustics, and visible and near-infrared spectroscopy. The DSC System integrates the DSC Probe, DSC software (v2023.10), and deployment equipment components to sense soil characteristics at a high vertical spatial resolution (mm scale) along in situ soil profiles up to a depth of 120 cm in about 60 s. The DSC Probe in situ proximal data are harmonized into a data cube providing vertical high-density knowledge associated with physical–chemical–biological soil conditions. In contrast, conventional ex situ soil samples derived from soil cores, soil pits, or surface samples analyzed using laboratory and other methods are bound by a substantially coarser spatial resolution and multiple compounding errors. Our objective was to investigate the effects of the mismatched scale between high-resolution in situ proximal sensor data and coarser-resolution ex situ soil laboratory measurements to develop soil prediction models. Our study was conducted in central California soil in almond orchards. We collected DSC sensor data and spatially co-located soil cores that were sliced into narrow layers for laboratory-based soil measurements. Partial Least Squares Regression (PLSR) cross-validation was used to compare the results of testing four data integration methods. Method A reduced the high-resolution sensor data to discrete values paired with layer-based soil laboratory measurements. Method B used stochastic distributions of sensor data paired with layer-based soil laboratory measurements. Method C allocated the same soil analytical data to each one of the high-resolution multi-sensor data within a soil layer. Method D linked the high-density multi-sensor soil data directly to crop responses (crop performance and behavior metrics), bypassing costly laboratory soil analysis. Overall, the soil models derived from Method C outperformed Methods A and B. Soil predictions derived using Method D were the most cost-effective for directly assessing soil–crop relationships, making this method well suited for industrial-scale precision agriculture applications. Full article

Soils play a central role in ecosystem functioning, and thus, mapped soil property information is indispensable to supporting sustainable land management. Digital Soil Mapping (DSM) provides a framework to spatially estimate soil properties. However, broad-scale DSM remains challenging because of non-purposively sampled soil data, large data volumes for processing extensive soil covariates, and high model complexities due to spatially varying soil–landscape relationships. This study presents a three-dimensional DSM framework for Switzerland, targeting the soil properties of clay content (Clay), organic carbon content (SOC), pH value (pH), and potential cation exchange capacity (CEC_pot). The DSM approach is based on machine learning and a comprehensive exploitation of soil and remote sensing data archives. Quantile Regression Forest was applied to link the soil sample data from a national soil data base with covariates derived from a LiDAR-based elevation model, from climate raster data, and from multispectral raster time series based on satellite imagery. The covariate set comprises spatially multiscale terrain attributes, climate patterns and their temporal variation, temporarily multiscale land use features, and spectral bare soil signatures. Soil data and predictions were evaluated with respect to different landcovers and depth intervals. All reference soil data sets were found to be spatially clustered towards croplands, showing an increasing sample density from lower to upper depth intervals. According to the R² value derived from independent data, the overall model accuracy amounts to 0.69 for Clay, 0.64 for SOC, 0.76 for pH, and 0.72 for CEC_pot. Reduced model accuracies were found to be accompanied by soil data sets showing limited sample sizes (e.g., CEC_pot), uneven statistical distributions (e.g., SOC), and low spatial sample densities (e.g., woodland subsoils). Multiscale terrain covariates were highly influential for all models; climate covariates were particularly important for the Clay model; multiscale land use covariates showed enhanced importance for modeling pH; and bare soil reflectance was a major driver in the SOC and CEC_pot models. Full article

Texture, soil organic matter (SOM), and soil depth (SoD) are crucial properties in forest management because they can supply spatial information on forest site productivity and guide fertilizer applications. However, soil properties possess an inherent uncertainty that must be mapped to enhance decision making in management applications. Most digital soil mapping predictions primarily concentrate on the mean of the distribution, often neglecting the estimation of local uncertainty in soil properties. Additionally, there is a noticeable scarcity of practical soil examples to demonstrate the prediction uncertainty for the benefit of forest managers. In this study, following a digital soil mapping (DSM) approach, a Quantile Regression Forest (QRF) model was developed to generate high-resolution maps and their uncertainty regarding the texture, SoD, and SOM, which were expressed as standard deviation (Sd) values. The results showed that the SOM (R² = 0.61, RMSE = 2.03% and with an average Sd = 50%), SoD (R² = 0.74 and RMSE = 19.4 cm), clay (R² = 0.63, RMSE = 10.5% and average Sd = 29%), silt (R² = 0.59, RMSE = 6.26% and average Sd = 33%), and sand content (R² = 0.55, RMSE = 9.49% and average Sd = 35%) were accurately estimated for forest plantations in central south Chile. A practical demonstration of precision fertilizer application, utilizing the predictive distribution of SOM, effectively showcased how uncertainty in soil attributes can be leveraged to benefit forest managers. This approach holds potential for optimizing resource allocation and maximizing economic benefits. Full article

The significant increase in hiking, wood extraction, and transportation activities exerts a notable impact on the environmental balance along trails and forest roads in the form of soil degradation. The aim of this study was to develop a Deformation Classification Model for the surface of a multi-use trail, as well as to calculate sediment deposition and generate a flood hazard map in a partially forested region. The eBee X mapping Unmanned Aerial Vehicle (UAV) equipped with the senseFly S.O.D.A. 3D camera and Real-Time Kinematic (RTK) technology flew over the study area of 149 ha in Northern Greece at an altitude of 120 m and achieved a high spatial resolution of 2.6 cm. The specific constellation of fixed-wing equipment makes the use of ground control points obsolete, compared to previous, in most cases polycopter-based, terrain deformation research. Employing the same methodology, two distinct classifications were applied, utilizing the Digital Surface Model (DSM) and Digital Elevation Model (DEM) for analysis. The Geolocation Errors and Statistics for Bundle Block Adjustment exhibited a high level of accuracy in the model, with the mean values for each of the three directions (X, Y, Z) being 0.000023 m, −0.000044 m, and 0.000177 m, respectively. The standard deviation of the error in each direction was 0.022535 m, 0.019567 m, and 0.020261 m, respectively. In addition, the Root Mean Square (RMS) error was estimated to be 0.022535 m, 0.019567 m, and 0.020262 m, respectively. A total of 20 and 30 altitude categories were defined at a 4 cm spatial resolution, each assigned specific ranges of values, respectively. The area of each altitude category was quantified in square meters (m²), while the volume of each category was measured in cubic meters (m³). The development of a Deformation Classification Model for the deck of a trail or forest road, coupled with the computation of earthworks and the generation of a flood hazards map, represents an efficient approach that can provide valuable support to forest managers during the planning phase or maintenance activities of hiking trails and forest roads. Full article

Readily available environmental covariates in current digital soil mapping usually do not indicate the spatial differences between deep soil attributes. This, to a large extent, leads to a decrease in the accuracy of 3D soil mapping with depth, which seriously affects the quality of soil information generated. This study tested the hypothesis that spatialized laboratory soil spectral information can be used as environmental covariates to improve the accuracy of 3D soil attribute mapping and proposed a new type of environmental covariable. In the first step, with soil-forming environmental covariates and independent soil profiles, laboratory vis-NIR spectral data of soil samples resampled into six bands in Anhui province, China, were spatially interpolated to generate spatial distributions of soil spectral measurements at multiple depths. In the second step, we constructed three sets of covariates using the laboratory soil spectral distribution maps at multiple depths: conventional soil-forming variables (C), conventional soil-forming variables plus satellite remote sensing wavebands (C+SRS) and conventional soil-forming variables plus spatialized laboratory soil spectral information (C+LSS). In the third step, we used the three sets of environmental covariates to develop random forest models for predicting soil attributes (pH; CEC, cation exchange capacity; Silt; SOC, soil organic carbon; TP, total phosphorus) at multiple depths. We compared the 3D soil mapping accuracies between these three sets of covariates based on another dataset of 132 soil profiles (collected in the 1980s). The results show that the use of spatialized laboratory soil spectral information as additional environmental covariates has a 50% improvement in prediction accuracy compared with that of only conventional covariates, and a 30% improvement in prediction accuracy compared with that of the satellite remote sensing wavebands as additional covariates. This indicates that spatialized laboratory soil spectral information can improve the accuracy of 3D digital soil mapping. Full article

Soil salinization is a worldwide problem that limits agricultural production. It is important to understand the salt stress tolerance ability of maize seedlings and explore the underlying related genetic resources. In this study, we used a high-throughput phenotyping platform with a 3D laser sensor (Planteye F500) to identify the digital biomass, plant height and normalized vegetation index under normal and saline conditions at multiple time points. The result revealed that a three-leaf period (T3) was identified as the key period for the phenotypic variation in maize seedlings under salt stress. Moreover, we mapped the salt-stress-related SNPs and identified candidate genes in the natural population via a genome-wide association study. A total of 44 candidate genes were annotated, including 26 candidate genes under normal conditions and 18 candidate genes under salt-stressed conditions. This study demonstrates the feasibility of using a high-throughput phenotyping platform to accurately, continuously quantify morphological traits of maize seedlings in different growing environments. And the phenotype and genetic information of this study provided a theoretical basis for the breeding of salt-resistant maize varieties and the study of salt-resistant genes. Full article

Deposition of corpses in the ground is the most common burial practice, which can allow interactions between polluting compounds and the soil, groundwater, and surface water, which may afterwards lead to negative environmental impacts and risks to public health. The risk of cemeteries contaminating groundwater is related to their location, the quantity of clothes, metals and adornments buried, and geographical, geological, hydrogeological, and climatic factors. Using the DRASTIC index and geographical information system (GIS) tools, the potential for groundwater contamination was investigated in eight cemeteries located in the Figueira da Foz region (Portugal), which are the main anthropogenic pollution sources in the area. Aquifer vulnerability was assessed through the development of thirteen site characteristic maps, seven thematic maps, and a DRASTIC index vulnerability map, using GIS operation tools. No studies were found on the development of vulnerability maps with this method and digital tools. Cemeteries UC2, UC4, UC5, UC6, UC7, and UC8 are located within the zones susceptible to recharge, with an average recharge rate of 254 mm/year. Cemeteries UC5, UC7, and UC8 are expected to develop a greater water-holding capacity. The water table depth is more vulnerable at UC6, varying between 9.1 m and 15.2 m. However, results show only a high vulnerability associated with the UC4 cemetery with the contributions T,C > R,S > I > A > D, which should be under an environmental monitoring program. The area surrounding UC4 is characterized by a water table depth ranging between 15.2 m to 22.9 m, mainly fine-grained sands in both the vadose zone and the aquifer media, Gleyic Solonchaks at the topsoil, very unfavorable slope (0–2%), and high hydraulic conductivity (>81.5 m/day). The sensitivity analysis shows that the topography, soil media, and aquifer media weights were the most effective in the vulnerability assessment. However, the highest contributions to index variation were made by hydraulic conductivity, net recharge, and soil media. This type of approach not only makes it possible to assess the vulnerability of groundwater to contamination from cemeteries but also allows the definition of environmental monitoring plans as well as provides the entities responsible for its management and surveillance with a methodology and tools for its continuous monitoring. Full article

Non-homogeneous soil’s high gravel content (also known as the “soil-rock dual structure”) may render it more prone to erosion and the significant development of gullies. In order to reveal the morphological characteristics and erosion rate of gullies in “soil-rock dual structure” areas, this study focused on the Shagou Reservoir basin in the Yimeng mountain area as the study area. Based on a complete digital orthophoto map (DOM, 0.03 m) and a digital elevation model (DEM, 0.03 m) acquired by an unmanned aerial vehicle (UAV), the researchers calculated the length (L), top width (TW), depth (D), area (A) and volume (V) of 19 gullies and built and optimized the volume estimation model. The DOM and the DEM were used to modify the morphological parameters of 43 gullies extracted from high-resolution remote sensing (RS) stereopair images (Worldview, 0.5 m), and the development and evolution of gully erosion were evaluated in large scale. The results showed that: (1) after correction, the average relative errors of parameters L, TW, D and A computed from the UAV data and the high-resolution RS stereopair image data fell below 0.005%; (2) the mean of TW/D was 5.20, i.e., the lateral erosion development of gullies far outweighed the downcutting erosion. The retrogressive erosion, lateral erosion and downcutting erosion rates of gullies were 0.01~0.83 m/a (averaged at 0.23 m/a), 0.01~0.68 m/a (averaged at 0.25 m/a) and 0.01~0.19 m/a (averaged at 0.09 m/a), respectively, between 2014 and 2021; (3) the volume-area (V-A) model for gullies is the optimal one (p < 0.01, R² = 0.944).A total of 90.7% of the gully volume was growing at an erosion rate of 0.42~399.39 m³/a and the total erosion rate of the gullies was 3181.56 m³/a from 2014 to 2021. These research findings can serve as a basis for the quantitative modeling of gully erosion in water-eroded locations with a large-dimension “soil-rock dual structure”. Full article