In agriculture, specifically livestock monitoring, drones' ability to track multiple targets is essential for advancing the field. However, limited computing resources and unpredictable drone movements often cause issues like ambiguous video frames, object obstructions, and size deviations. These inconsistencies reduce tracking accuracy, making traditional algorithms inadequate for handling drone footage. This study introduces an enhanced deep learning-based multi-target drone tracker framework that enables real-time processing. The proposed method combines object detection and tracking by leveraging consecutive frame pairs to extract and share features, enhancing computational efficiency. It employs diverse loss functions to address class and sample distribution imbalances and includes a composite deblurring module to enhance detection accuracy. Object association utilizes a dual regress bounding box technique, aiding in object identification verification and predictive motion. Live tracking is achieved by predicting object locations in subsequent frames, enabling real-time tracking. Evaluation against leading benchmarks shows that the system improves precision and speed, achieving a 4.3 % increase in Multi-Object Tracking Accuracy (MOTA) and a 7.7 % boost in F1 score.

A complete understanding of groundwater circulation as well as the transport and distribution of solutes in arid-semiarid basin regions is a prerequisite for the safe use of groundwater resources. The distributions of the groundwater flow systema and solutes are affected by the basin morphology, lithology, and climate variations; therefore, they can change over geologic time. In this study, we performed a case study of the Qaidam Basin in the northeastern Tibetan Plateau, in which we utilized reactive solute transport simulations to build a numerical model in TOUGHREACT for a typical section of the eastern Qaidam Basin since the last glacial period. The results show that the groundwater in the eastern Qaidam Basin developed into a three-level groundwater flow system and that the seepage velocity of the local water flow system is significantly higher than that of the intermediate and regional water flow systems. Although groundwater in the discharge region has been continuously concentrated and enriched since the last glacial period, the distributions of the groundwater flow system and solutes have been greatly affected by climate variations. During warm periods, the centres of groundwater discharge and solute concentration shifted to areas with more groundwater recharge; in contrast, both centres shifted to the central basin during drought periods. The groundwater in the basin mainly contains Na⁺and Cl⁻ions, which vary significantly from the recharge region to the discharge region. Evaporation of groundwater results in increases in the concentrations of most of the components except HCO₃⁻. The groundwater in the discharge region is currently in the stage of carbonate precipitation and is far from gypsum and halite precipitation.

An ecological security pattern is an essential natural solution for promoting the coordinated and sustainable development of urban growth and the ecological environment within a given region. This study aims to develop a comprehensive ecological security pattern (ESP) for the Jinjiang watershed by integrating multiple ecological service functions: ecological regulation, eco-product provisions, and human settlement services. Using a well-established methodology of “source identification—resistance surface construction—ecological corridor extraction”, we identified 322 ecological source areas and delineated 321 ecological corridors spanning 1941.67 km, with key corridors connecting ecological regulation sources and ordinary corridors primarily linking eco-product provisions and human settlement sources. Our results show that the northern hilly regions are significant for ecological regulation, Anxi County is key for eco-product provisions, and human settlement services are scattered across urban areas and river valleys. This integration enhances regional ecological sustainability and provides a model for similar coastal regions globally. Our findings offer scientific guidance for balancing urban growth with ecological preservation, contributing to sustainable development worldwide.

Rivers are an important source of water in humid regions, but their availability is greatly limited by water chemistry. In order to accurately identify the changes in river water chemical composition, the compositional analysis method (CoDA) is proposed from the perspective of compositional data analysis theory, which considers the geochemical riverine system as a whole and detects the compositional changes of the entire watershed. The basic data analysis is carried out by traditional analysis methods, and the results show that the hydrochemical characteristics of different sections of the basin have significant features. The water chemistry of Dongxi River is of the HCO₃⁻ Ca type. The water of the Xixi River shows a gradual evolution from the HCO₃-Ca type and high SO₄²⁻ content in the upper reaches to the Cl-Ca type in the lower reaches. The hydrochemistry of river water in the watershed is mainly affected by rock weathering leaching (PC1) and agricultural and domestic pollutant discharge (PC2), with a contribution rate of 48.4% and 19.7%, respectively. Rock weathering, mining, and agricultural pollution are the main factors affecting the chemical composition of river water in different regions. The spatial composition of a single sample at different scales is monitored by the Mahalanobis distance approach in an iterative manner to minimize the influence of a single anomaly on the composition center. The results show that the main reasons for the change in river water chemical composition along the Xixi River are attributed to mine pollution, domestic pollution, and tea plantation and that along the Dongxi River is caused by domestic pollution. The hydrochemical composition changes after the confluence of the Xixi River and Dongxi River are mainly affected by human activities and seawater in urban areas. This research could provide new perspectives and methods for detecting the influences of human and natural factors on the hydrochemistry of river water in humid regions worldwide.

Investigating the spatial distribution of chromium (Cr) in soil is essential for understanding Cr pollution and accurately assessing associated environmental risks. However, field sampling is challenging due to limited sampling points, and the spatial distribution of Cr is affected by multiple complex environmental covariates, thereby restricting model development and prediction accuracy. This study selected the Chizhou–Xuancheng border area in southern Anhui Province as the research region and collected 2035 data points. Machine learning models, including AdaBoost, GBDT, XGBoost, and MLP, were employed to predict Cr concentrations in conjunction with environmental covariates. To address the challenges of sparse sampling data and complex data relationships for Cr prediction, the PHMS-Transformer model is proposed. Featuring a shallow encoder design, configurable pooling strategies, and a lightweight structure, the model significantly reduces the number of parameters and alleviates overfitting under sparse sampling conditions, while the incorporation of multi-head self-attention mechanisms captures complex nonlinear relationships among multi-source environmental variables relevant to Cr. To further enhance model interpretability for Cr prediction, the SHAP model was applied to identify key factors influencing Cr distribution. Comprehensive comparisons indicate that the PHMS-Transformer model achieves superior performance in predicting Cr, demonstrating high accuracy and generalization capability, with clear advantages over traditional methods. These findings offer valuable insights for soil environmental protection and Cr pollution control and possess significant theoretical and practical implications. Soil Cr pollution represents a global environmental challenge, where achieving accurate predictions for Cr is particularly crucial yet difficult in regions with constrained data accessibility. The lightweight, high-precision, and interpretable PHMS-Transformer framework proposed in this study provides an effective technical solution to the widespread challenges of sample sparsity and model complexity inherent in predicting the spatial distribution of soil Cr globally. Therefore, this work offers significant reference value for advancing global soil environmental risk assessment and Cr pollution remediation efforts.