With global climate change and the deterioration of the ecological environment, the safety of hydraulic engineering faces severe challenges, among which soil-dwelling termite damage has become an issue that cannot be ignored. Reservoirs and embankments in China, primarily composed of earth and rocks, are often affected by soil-dwelling termites, such as Odontotermes formosanus and Macrotermes barneyi. Identifying soil-dwelling termite damage is crucial for implementing monitoring, early warning, and control strategies. This study developed an improved YOLOv8 model, named MCD-YOLOv8, for identifying traces of soil-dwelling termite activity, based on the Monte Carlo random sampling algorithm and a lightweight module. The Monte Carlo attention (MCA) module was introduced in the backbone part to generate attention maps through random sampling pooling operations, addressing cross-scale issues and improving the recognition accuracy of small targets. A lightweight module, known as dimension-aware selective integration (DASI), was added in the neck part to reduce computation time and memory consumption, enhancing detection accuracy and speed. The model was verified using a dataset of 2096 images from the termite damage survey in hydraulic engineering within Hubei Province in 2024, along with images captured by drone. The results showed that the improved YOLOv8 model outperformed four traditional or enhanced models in terms of precision and mean average precision for detecting soil-dwelling termite damage, while also exhibiting fewer parameters, reduced redundancy in detection boxes, and improved accuracy in detecting small targets. Specifically, the MCD-YOLOv8 model achieved increases in precision and mean average precision of 6.4% and 2.4%, respectively, compared to the YOLOv8 model, while simultaneously reducing the number of parameters by 105,320. The developed model is suitable for the intelligent identification of termite damage in complex environments, thereby enhancing the intelligent monitoring of termite activity and providing strong technical support for the development of termite control technologies.

Mountain permafrost extends over a vast area throughout the Chilean and Argentinean Andes, making it a key component of these mountain ecosystems. To develop an overview of the current state of knowledge on southern Andean permafrost, it is essential to outline appropriate research strategies in a warmer climate context. Based on a comprehensive review of existing literature, this work identifies eight main research themes on mountain permafrost in the Chilean and Argentinean Andes: paleoenvironmental reconstructions, permafrost-derived landforms inventories, permafrost distribution models, internal structure analysis, hydrogeochemistry, permafrost dynamics, geological hazards, and transitional landscape studies. This extensive review work also highlights key debates concerning the potential of permafrost as a water resource and the factors influencing its distribution. Furthermore, we identified several challenges the scientific community must address to gain a deeper understanding of mountain permafrost dynamics. Among these challenges, we suggest tackling the need to broaden spatial focus, along with the use of emerging technologies and methodologies. Additionally, we emphasize the importance of developing interdisciplinary approaches to effectively identify the impacts of climate change on mountain permafrost. Such efforts are essential for adequately preparing scientists, institutional entities, and society to address future scenarios.

在冰湖编目工作中,从海量遥感数据快速准确获取冰湖边界具有重要意义,发展基于遥感数据的冰湖边界自动化提取方法是关键。本研究改进了基于YOLOv5-Seg网络的实例分割模型,并应用于山地冰湖自动化提取。结果显示,使用坐标注意力机制（Coordinate-Attention,CA）,提高网络对冰湖目标的关注程度;在原始3个检测层的基础上添加小目标检测层,增强网络对小面积冰湖检测能力;修改上采样方法为转置卷积,解决了最近邻上采样丢失特征问题。改进的YOLOv5-Seg网络比原始网络平均精度提升2.7%,达到75.1%,比目前其他主流算法精度高10%。利用改进的YOLOv5-Seg网络的实例分割模型和Sentinel-2卫星影像,发现2022年兴都库什—喀喇昆仑—喜马拉雅地区（HKH）,共有10 668个冰湖正例,共计768.3 km2。该研究通过深度卷积神经网络和多源遥感数据,为大地理区域的自动冰湖制图提供了技术支持。

在冰湖编目工作中,从海量遥感数据快速准确获取冰湖边界具有重要意义,发展基于遥感数据的冰湖边界自动化提取方法是关键。本研究改进了基于YOLOv5-Seg网络的实例分割模型,并应用于山地冰湖自动化提取。结果显示,使用坐标注意力机制（Coordinate-Attention,CA）,提高网络对冰湖目标的关注程度;在原始3个检测层的基础上添加小目标检测层,增强网络对小面积冰湖检测能力;修改上采样方法为转置卷积,解决了最近邻上采样丢失特征问题。改进的YOLOv5-Seg网络比原始网络平均精度提升2.7%,达到75.1%,比目前其他主流算法精度高10%。利用改进的YOLOv5-Seg网络的实例分割模型和Sentinel-2卫星影像,发现2022年兴都库什—喀喇昆仑—喜马拉雅地区（HKH）,共有10 668个冰湖正例,共计768.3 km2。该研究通过深度卷积神经网络和多源遥感数据,为大地理区域的自动冰湖制图提供了技术支持。

Forests are essential to our planet's well-being, playing a vital role in climate regulation, biodiversity preservation, and soil protection, thus serving as a cornerstone of our global ecosystem. The threat posed by forest fires highlights the critical need for early detection systems, which are indispensable tools in safeguarding ecosystems, livelihoods, and communities from devastating destruction. In combating forest fires, a range of techniques is employed for efficient early detection. Notably, the combination of drones with artificial intelligence, particularly deep learning, holds significant promise in this regard. Image segmentation emerges as a versatile method, involving the partitioning of images into multiple segments to simplify representation, and it leverages deep learning for fire detection, continuous monitoring of high-risk areas, and precise damage assessment. This study provides a comprehensive examination of recent advancements in semantic segmentation based on deep learning, with a specific focus on Mask R-CNN (Mask Region Convolutional Neural Network) and YOLO (You Only Look Once) v5, v7, and v8 variants. The emphasis is placed on their relevance in forest fire monitoring, utilizing drones equipped with high-resolution cameras.

Ground ice distribution and abundance have wide-ranging effects on periglacial environments and possible impacts on climate change scenarios. In contrast, very few studies measure ground ice in the High Arctic, especially in polar deserts and where coarse surficial material complicates coring operations. Ground ice volumes and cryostructures were determined for eight sites in a polar desert, near Resolute Bay, Nunavut, chosen for their hydrogeomorphic classification. Dry, unvegetated polar desert sites exhibited ice content close to soil porosity, with a <45 cm thick ice-enriched transition zone. In wetland sites, suspended cryostructures and ice dominated cryofacies (ice content at least 2x soil porosity values) were prevalent in the upper similar to 2 m of permafrost. Average ground ice saturation at those locations exceeded porosity values by a factor between 1.8 and 20.1 and by up to two orders of magnitude at the similar to 10 cm vertical scale. Sites with the highest ice contents were historically submerged wetlands with a history of sediment supply, sustained water availability, and syngenetic and quasi-syngenetic permafrost aggradation. Ice enrichment in those environments were mainly caused by the strong upward freezing potential beneath the thaw front, which, combined with abundant water supply, caused ice aggradation and frost heaving to form lithalsa plateaus. Most of the sites already expressed cryostratigraphic evidence of permafrost degradation. Permafrost degradation carries important ecological ramifications, as wetland locations are the most productive, life-supporting oases in the otherwise relatively barren landscape, carrying essential functions linked with hydrological processes and nutrient and contaminant cycling.

高原高寒地区环境复杂、昼夜温差大，对铁路建设与运营产生不良影响，容易发生由于钢轨内部伤损引起的断裂事故。研究基于YOLOv5的超声波图像识别技术在青藏铁路钢轨探伤检测中的应用，特别是针对轨头核伤、轨面鱼鳞伤等常见伤损类型进行检测。通过智能钢轨探伤仪采集高寒地段钢轨数据，以YOLOv5方法对数据集进行整理、处理和模型训练，精准识别和定位轨头核伤、轨面鱼鳞伤等损伤。研究表明，基于YOLOv5的模型在识别和定位各类钢轨损伤方面具有很高的准确性和实时性，可以同时进行目标检测和类别分类，并能在保持较高准确度的同时实现快速检测。提供一种新的、更有效的钢轨探伤检测数据分析方法，有助于提高铁路安全和运营效率。

高原高寒地区环境复杂、昼夜温差大，对铁路建设与运营产生不良影响，容易发生由于钢轨内部伤损引起的断裂事故。研究基于YOLOv5的超声波图像识别技术在青藏铁路钢轨探伤检测中的应用，特别是针对轨头核伤、轨面鱼鳞伤等常见伤损类型进行检测。通过智能钢轨探伤仪采集高寒地段钢轨数据，以YOLOv5方法对数据集进行整理、处理和模型训练，精准识别和定位轨头核伤、轨面鱼鳞伤等损伤。研究表明，基于YOLOv5的模型在识别和定位各类钢轨损伤方面具有很高的准确性和实时性，可以同时进行目标检测和类别分类，并能在保持较高准确度的同时实现快速检测。提供一种新的、更有效的钢轨探伤检测数据分析方法，有助于提高铁路安全和运营效率。

Permafrost in the NE European Russian Arctic is suffering from some of the highest degradation rates in the world. The rising mean annual air temperature causes warming permafrost, the increase in the active layer thickness (ALT), and the reduction of the permafrost extent. These phenomena represent a serious risk for infrastructures and human activities. ALT characterization is important to estimate the degree of permafrost degradation. We used a multidisciplinary approach to investigate the ALT distribution in the Khanovey railway station area (close to Vorkuta, Arctic Russia), where thaw subsidence leads to railroad vertical deformations up to 2.5 cm/year. Geocryological surveys, including vegetation analysis and underground temperature measurements, together with the faster and less invasive electrical resistivity tomography (ERT) geophysical method, were used to investigate the frozen/unfrozen ground settings between the railroad and the Vorkuta River. Borehole stratigraphy and landscape microzonation indicated a massive prevalence of clay and silty clay sediments at shallow depths in this area. The complex refractive index method (CRIM) was used to integrate and quantitatively validate the results. The data analysis showed landscape heterogeneity and maximum ALT and permafrost thickness values of about 7 and 50 m, respectively. The active layer was characterized by resistivity values ranging from about 30 to 100 omega m, whereas the underlying permafrost resistivity exceeded 200 omega m, up to a maximum of about 10 k omega m. In the active layer, there was a coexistence of frozen and unfrozen unconsolidated sediments, where the ice content estimated using the CRIM ranged from about 0.3 - 0.4 to 0.9. Moreover, the transition zone between the active layer base and the permafrost table, whose resistivity values ranged from 100 to 200 omega m for this kind of sediments, showed ice contents ranging from 0.9 to 1.0. Taliks were present in some depressions of the study area, characterized by minimum resistivity values lower than 10 omega m. This thermokarst activity was more active close to the railroad because of the absence of insulating vegetation. This study contributes to better understanding of the spatial variability of cryological conditions, and the result is helpful in addressing engineering solutions for the stability of the railway.

Considering different physicographical territory under changing climate conditions, a quantitative technique is presented for estimating the changes in bearing capacity of the permafrost foundations. The results showed an increase in the permafrost temperature over 30 years (1960-1990) due to climate warming. This led to a decrease in the bearing capacity foundations in the north of Western Siberia, and in some regions the reduction was up to 45%. The predicted climate warming may lead to a further decrease in the bearing capacity of the foundations built on the principle of permafrost construction, which will lead to an increase in the number of deformations of buildings and structures and may adversely affect the development of the region's infrastructure.