Natural and planted forests, covering approximately 31% of the Earth’s land area, are crucial for global ecosystems, providing essential services such as regulating the water cycle, soil conservation, carbon storage, and biodiversity preservation. However, traditional forest mapping and monitoring methods are often costly and limited in scale, highlighting the need to develop innovative approaches for tree detection that can enhance forest management. In this study, we present a new dataset for tree detection, VHRTrees, derived from very high-resolution RGB satellite images. This dataset includes approximately 26,000 tree boundaries derived from 1,496 image patches of different geographical regions, representing various topographic and climatic conditions. We implemented various object detection algorithms to evaluate the performance of different methods, propose the best experimental configurations, and generate a benchmark analysis for further studies. We conducted our experiments with different variants and hyperparameter settings of the YOLOv5, YOLOv7, YOLOv8, and YOLOv9 models. Results from extensive experiments indicate that, increasing network resolution and batch size led to higher precision and recall in tree detection. YOLOv8m, optimized with Auto, achieved the highest F1-score (0.932) and mean Average Precision (mAP)@0.50 Intersection over Union threshold (0.934), although some other configurations showed higher mAP@0.50:0.95. These findings underscore the effectiveness of You Only Look Once (YOLO)-based object detection algorithms for real-time forest monitoring applications, offering a cost-effective and accurate solution for tree detection using RGB satellite imagery. The VHRTrees dataset, related source codes, and pretrained models are available at https://github.com/RSandAI/VHRTrees.

Wheat is a key staple crop globally, essential for food security and sustainable agricultural development. The results of this study highlight how innovative monitoring techniques, such as UAV-based multispectral imaging, can significantly improve agricultural practices by providing precise, real-time data on crop growth. This study utilized unmanned aerial vehicle (UAV)-based remote sensing technology at the wheat experimental field of the Hebei Academy of Agriculture and Forestry Sciences to capture the dynamic growth characteristics of wheat using multispectral data, aiming to explore efficient and precise monitoring and management strategies for wheat. A UAV equipped with multispectral sensors was employed to collect high-resolution imagery at five critical growth stages of wheat: tillering, jointing, booting, flowering, and ripening. The data covered four key spectral bands: green (560 nm), red (650 nm), red-edge (730 nm), and near-infrared (840 nm). Combined with ground-truth measurements, such as chlorophyll content and plant height, 21 vegetation indices were analyzed for their nonlinear relationships with wheat growth parameters. Statistical analyses, including Pearson’s correlation and stepwise regression, were used to identify the most effective indices for monitoring wheat growth. The Normalized Difference Red-Edge Index (NDRE) and the Triangular Vegetation Index (TVI) were selected based on their superior performance in predicting wheat growth parameters, as demonstrated by their high correlation coefficients and predictive accuracy. A random forest model was developed to comprehensively evaluate the application potential of multispectral data in wheat growth monitoring. The results demonstrated that the NDRE and TVI indices were the most effective indices for monitoring wheat growth. The random forest model exhibited superior predictive accuracy, with a mean squared error (MSE) significantly lower than that of traditional regression models, particularly during the flowering and ripening stages, where the prediction error for plant height was less than 1.01 cm. Furthermore, dynamic analyses of UAV imagery effectively identified abnormal field areas, such as regions experiencing water stress or disease, providing a scientific basis for precision agricultural interventions. This study highlights the potential of UAV-based remote sensing technology in monitoring wheat growth, addressing the research gap in systematic full-cycle analysis of wheat. It also offers a novel technological pathway for optimizing agricultural resource management and improving crop yields. These findings are expected to advance intelligent agricultural production and accelerate the implementation of precision agriculture.

Incomplete environmental datasets pose significant challenges in developing accurate predictive models, particularly in hydrological research. This study addresses data missingness by investigating gap imputation methodologies for datasets with 5–20% data absence, focusing on the Mureș River in Romania. Utilizing a novel approach, we applied various imputation techniques, including the ratio method, Kalman filtering, and machine learning algorithms (XGBoost, Gradient Boosting, Random Forest and CatBoost), while developing an innovative self-assessment metric for evaluating imputation performance without relying on external reference data. Through systematic analysis of hydrological station data from four monitoring points, we artificially introduced data gaps to rigorously test method applicability. The research demonstrates the feasibility of constructing a robust self-evaluation framework for selecting optimal imputation techniques, potentially enhancing data reliability and analytical precision in environmental and geospatial research. Our findings contribute a structured methodology for addressing data incompleteness, offering researchers a quantitative approach to improving dataset integrity and predictive modeling in complex environmental systems.

Soil salinization significantly impacts global agricultural productivity, contributing to desertification and land degradation; thus, rapid regional monitoring of soil salinization is crucial for agricultural production and sustainable management. With advancements in artificial intelligence, the efficiency and precision of deep learning classification models applied to remote sensing imagery have been demonstrated. Given the limited feature learning capability of traditional machine learning, this study introduces an innovative deep fusion U-Net model called MSA-U-Net (Multi-Source Attention U-Net) incorporating a Convolutional Block Attention Module (CBAM) within the skip connections to improve feature extraction and fusion. A salinized soil classification dataset was developed by combining spectral indices obtained from Landsat-8 Operational Land Imager (OLI) data and polarimetric scattering features extracted from RADARSAT-2 data using polarization target decomposition. To select optimal features, the Boruta algorithm was employed to rank features, selecting the top eight features to construct a multispectral (MS) dataset, a synthetic aperture radar (SAR) dataset, and an MS + SAR dataset. Furthermore, Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (KNN), and deep learning methods including U-Net and MSA-U-Net were employed to identify the different degrees of salinized soil. The results indicated that the MS + SAR dataset outperformed the MS dataset, with the inclusion of the SAR band resulting in an Overall Accuracy (OA) increase of 1.94–7.77%. Moreover, the MS + SAR MSA-U-Net, in comparison to traditional machine learning methods and the baseline model, improved the OA and Kappa coefficient by 8.24% to 12.55% and 0.08 to 0.15, respectively. The results demonstrate that the MSA-U-Net outperformed traditional models, indicating the potential of integrating multi-source data with deep learning techniques for monitoring soil salinity.

The nature of wildlife management throughout the world is changing. The increase in the world’s human population has been accompanied by a rapid expansion of agricultural and urban areas and infrastructures, especially road and railway networks. Worldwide, wildlife habitats are being transformed and fragmented by human activities, and the behavior of several species has changed as a result of human activities. Some species have adapted easily to urban or peri–urban habitats and take advantage of the new resources available. These data provide the context for why human–wildlife interactions are increasing. At the 30th International Union of Game Biologists Congress held in Barcelona in early September 2011, in addition to two plenary presentations, 52 authors from 12 different countries and three continents presented 15 papers in the Interactions of Humans and Wildlife Session, three of which are included in this volume. To some extent, all the papers reflected the inherent difficulty in solving the complex problems caused either by rapidly increasing species that begin to inhabit urban and agricultural areas in numbers not seen previously (e.g. coyotes, Canis latrans, inhabiting big cities; wild boar, Sus scrofa, across western Europe; wood pigeons, Columba palumbus, in France), or species whose populations are threatened by human activities (e.g., Eurasian Lynx, Lynx lynx, in the Czech Republic). Some papers addressed the contentious issue of predator control (e.g., gamebirds in Great Britain), while others presented data regarding how human activities influenced animal behavior (e.g., pink footed geese, Anser brachyrhynchus; and red deer, Cervus elaphus, in Germany). The papers presented at the congress show how human activities affect the distributions and dynamics of wildlife populations and also change the behavior of some species. Wildlife causes social and economic conflicts by damaging agricultural and forest resources, bringing about traffic collisions, and creating problems for residents in urban areas; while many are increasingly distant from nature and may not accept the presence of wildlife others may actively encourage the presence of wild animals. The first paper in this volume, by Cahill et al. (2012), analyzes the management challenges of the increasing abundance of wild boar in the peri–urban area of Barcelona. This conflict has arisen in other large cities in Europe and elsewhere. The presence of the species causes problems for many residents, to such an extent that it is considered a pest in these areas. Wild boar habituation has not only been facilitated by population expansion, but also by the attitudes of some citizens who encourage their presence by direct feeding. This leads to wild boar behavior modification and also promotes an increase in the fertility rate of habituated females, which are significantly heavier than non–habituated females. Public attitudes regarding the species and harvesting methods (at present most specimens are removed by live capture and subsequently sacrificed) are highlighted as one of the key factors in the management of the conflict. The second paper provides an example of how the distribution of irrigated croplands influences wild boar roadkills in NW Spain (Colino–Rabanal et al., 2012). By modeling the spatial distribution of wild boar collisions with vehicles and using generalized additive models based on GIS, the authors show that the number of roadkills is higher in maize croplands than in forested areas. This factor is the main explanatory variable in the model. The paper provides an excellent example of how the synergies of diverse human elements in the landscape (maize croplands and roads in this case) affect the location and dimensions of these types of conflicts. The third and final paper, by Belotti et al. (2012), addresses the effects of tourism on Eurasian lynx movements and prey usage at Šumava National Park in the Czech Republic. The monitoring of 5 GPS–collared lynxes and analyses of data regarding habitat features suggests that human disturbance (proximity of roads and tourist trails) can modify the presence of lynxes during the day close to the site where they have hidden a prey item, such as an ungulate, that can provide them with food for several days. In such cases, adequate management of tourism development must involve a commitment to species conservation. The analyses and understanding of all these phenomena and the design of successful wildlife management strategies and techniques used to mitigate the conflicts require a good knowledge base that considers information both about wildlife and human attitudes. The papers presented stress the importance of spatial analyses of the interactions and their relationship with landscape features and the location of human activities. Species distribution and abundance are related to important habitat variables such as provision of shelter, food, comfortable spaces, and an appropriate climate. Therefore, it is essential to analyze these data adequately to predict where conflicts are most likely to arise and to design successful mitigation strategies. The second key factor for adequate management of human–wildlife interactions is to monitor system change. An analysis of the variety of data on population dynamics, hunting, wildlife collisions, and wildlife presence in urban areas would provide a basis for adaptive management. In this respect, in the plenary session, Steve Redpath mentioned the importance of the wildlife biologist’s attitude when interpreting and drawing conclusions from recorded data and stressed the importance of conducting clear, relevant, and transparent science for participants involved in the management decision process, which often involves a high number of stakeholders. All of the papers addressing the problems associated with human wildlife interactions were characterized by a common theme. Regardless of the specific nature of the problem, the public was generally divided on how the problem should be addressed. A particularly sensitive theme was that of population control methods, especially when conflicts are located in peri–urban areas. Several presenters acknowledged that public participation was necessary if a solution was to be reached. Some suggested, as have other authors (Heydon et al., 2010), that a legislative framework may be needed to reconcile human and wildlife interests. However, each problem that was presented appeared to involve multiple stakeholders with different opinions. Solving these kinds of problems is not trivial. Social factors strongly influence perceptions of human–wildlife conflicts but the methods used to mitigate these conflicts often take into account technical aspects but not people’s attitudes. A new, more innovative and interdisciplinary approach to mitigation is needed to allow us 'to move from conflict towards coexistence' (Dickman, 2010). Other authors also mentioned the importance of planning interventions that optimize the participation of experts, policy makers, and affected communities and include the explicit, systematic, and participatory evaluation of the costs and benefits of alternative interventions (Treves et al., 2009). One technique that has been used to solve problems like these is termed Structured Decision Making (SDM). This technique was developed by the U.S. Geological Survey and the U.S. Fish and Wildlife Service. As described by Runge et al. (2009), the process is 'a formal application of common sense for situations too complex for the informal use of common sense', and provides a rational framework and techniques to aid in prescriptive decision making. Fundamentally, the process entails defining a problem, deciding upon the objectives, considering the alternative actions and the consequences for each, using the available science to develop a model (the plan), and then making the decision how to implement (Runge et al., 2009). Although complex, SDM uses a facilitator to guide stakeholders through the process to reach a mutually agreed–upon plan of action. It is clear that human–wildlife interactions are inherently complex because many stakeholders are usually involved. A rational approach that incorporates all interested parties would seem to be a productive way of solving these kinds of problems.