With the rapid accumulation of marine big data and the robust development of Artificial Intelligence (AI) technology, intelligent marine forecasting has shown greater precision and efficiency in this new era. Marine data can be categorized into point- and field-observation data based on the observation methods, providing foundational support for marine forecasting. Marine forecasting methods can be divided into three main types based on the characteristics of the dynamic marine processes and phenomena: point-to-point, field-to-point, and field-to-field forecasting. These forecasting approaches not only cover a variety of marine phenomena but also address different forecasting requirements. Through a case analysis, this study specifically introduces intelligent forecasting models and results for point-to-point internal solitary wave forecasting, field-to-point El Niño-Southern Oscillation (ENSO) forecasting, and field-to-field phenomena such as mesoscale eddies and sea ice. Finally, it explores the development directions for intelligent marine forecasting in the context of big data, suggesting that enhancing the integration of data-driven methods with physical mechanisms can improve forecast accuracy and real-time responsiveness, thereby providing technical support for marine environmental monitoring, disaster warning, and the sustainable use of marine resources.

The study of Source-to-Sink systems is an important field of research focused on understanding the entire process of material transport from source areas like mountain ranges or other landforms to sink areas like river basins, lakes, and oceans. This process entails weathering of the parent rock, erosion of materials, transportation via various agents (such as wind, water, or ice), and eventual deposition at sink locations. Analyzing this system reveals dynamic surface changes, material cycling mechanisms, and how these processes adapt to environmental shifts over time. Understanding these complex processes is crucial for a variety of scientific fields, including geomorphology, environmental science, and natural resource management; however, the traditional methods such as field observations and laboratory analyses, Have their own set of challenges. Data availability, low spatiotemporal resolution, and ambiguity in interpretation make it difficult to capture the rapid and dynamic changes occurring in natural systems. Furthermore, these methods are not ideally suited for analyzing long-term evolutionary processes or large-scale systems. Consequently, numerical modeling has emerged as an essential tool studying source-to-sink systems, addressing these traditional limitations by simulating complex processes over varying spatial and temporal scales. They offer more quantitative insights into the dynamics of erosion, transport, and deposition under different environmental conditions.This paper reviews five key numerical tools commonly used in source-to-sink research: Dionisos, SEDSIM, Landlab, goSPL, and Delft3D. Each tool has specific advantages that render it suitable for various research purposes. Dionisos, for instance, excels at modeling large-scale, long-term basin-filling processes though it is less effective for simulating small-scale, dynamic changes. SEDSIM, based on hydrodynamic equations, produces highly accurate results for clastic sedimentary processes, but tends to be slower and more focused on specific types of sediment. LandLab is highly customizable and capable of multi-process simulations; although, it requires advanced programming skills. goSPL handles global-scale high-resolution simulations effectively, despite struggling with localized phenomena and requiring significant computational resources. Delft3D is ideal for small-scale, fine-detail simulations, particularly in coastal, riverine, and lacustrine environments, although it faces challenges in large-scale applications. With ongoing advances in computational power and algorithms, future advancements in source-to-sink modeling are expected. The integration of big data and AI will likely enhance the accuracy of predictions, facilitate multidisciplinary integration, and drive the intelligent evolution of the field.

Over more than 50 years of continuous research and technological innovation, Fengyun Meteorological Satellite System has achieved significant progress. 21 Fengyun satellites have been launched. Currently, eight of these satellites operate stably in orbit, forming a comprehensive observation system that includes geostationary orbit and sun-synchronous polar orbit satellites. By reviewing the development history and current status of Fengyun meteorological satellites and remote sensing instruments; the effectiveness of ground segments in data reception, processing, and operation; and the construction and service of application systems, the technical capabilities of Fengyun meteorological satellites, their ground segments, and application systems were comprehensively analyzed. Through comparative analysis with major countries around the world in terms of meteorological satellite network observations, remote sensing instrument technology, and ground segment operation capabilities, it was found that the Fengyun Meteorological Satellites not only have a complete orbit layout and remote sensing instrument configuration, but their remote sensing instrument detection capability has reached the advanced international level, although some performance indicators still have spcace for improvement. Ground segments have established efficient data reception, processing, and service processes with advanced data preprocessing technology and sub-pixel-level geolocation accuracy. The radiometric calibration accuracy is 3% in the visible and near infrared channels and 0.2 K in the infrared channels. In addition, the Fengyun Meteorological Satellite System has established a comprehensive and complete quantitative product system for atmospheric, land, marine, and space weather, and has established China Radiometric Calibration Sites for Chinese remote sensing satellites, and carried out validation of the remote sensing products. Fengyun satellite data have been widely used in various fields, such as weather forecasting, climate change research, ecological environment monitoring, and natural disaster warning, and their application level continues to advance. In the future, the Fengyun meteorological satellite observation system will aim to evolve towards establishing a hybrid-architecture space observation system, achieving comprehensive and precise perception of observation elements, enabling intelligent and efficient operation of satellite-ground systems, integrating emerging technologies in data processing, expanding remote sensing application scenarios, and fostering international cooperation and sharing.

In the summer of 2022, the Yangtze River Basin experienced unprecedented heat waves, drawing considerable attention from the scientific community. Affected by over a month of record-breaking high temperatures and droughts, this extreme event not only caused escalating losses to human health, the economy, and the environment, but also exacerbated food insecurity and hindered sustainable development. Therefore, a more comprehensive understanding of extreme heat in the Yangtze River Basin during the summer of 2022 is essential to identify the drivers of extreme event variability under global warming, assess the impacts of human activity and natural variability, and evaluate potential climate risks. This study first reviews the main characteristics, formation mechanisms, and causes of the extreme heat in the Yangtze River Basin in the summer of 2022, and further summarizes the research progress on the event over the past three years. The results showed that the 2022 summer high temperature in the Yangtze River Basin was a rare extreme heat event. Its occurrence was primarily driven by atmospheric circulation anomalies related to the western Pacific subtropical high and the South Asian high, the triple La Niña phenomenon, the Atlantic and Indian SST forcing, and land-atmosphere feedback mechanism (e.g., soil moisture and air temperature). In addition to natural variability, human activity is the dominant factor influencing heat extremes. Without anthropogenic forcing, such extremes would have been highly unlikely. Such rare heat waves are projected to become more frequent under ongoing global warming. Finally, the paper highlights key research challenges and knowledge gaps associated with extreme heat events.

As one of the most significant cryospheric landforms that respond to climate warming in permafrost regions, thermokarst lakes profoundly influence ecological changes, regional hydrological cycles, and biogeochemical processes while compromising the stability of permafrost engineering. This study reviews recent advances in the formation and evolution of thermokarst lakes, their hydrological cycles, heat transfer, ecological and environmental effects, and engineering impacts across northern hemisphere permafrost regions. Research indicates that in the discontinuous permafrost zones of the Arctic, lake and pond areas show a predominantly decreasing trend, whereas, in continuous permafrost zones, both expansion and shrinkage are observed. On the Qinghai-Tibet Plateau, climate warming and increased precipitation have led to the rapid formation and expansion of thermokarst lakes. The evolution of these lakes, coupled with hydrological cycling and thermal effects, alters the physicochemical properties of the surrounding soils, influences hydrothermal dynamics in alpine ecosystems, and reduces the stability of adjacent permafrost engineering structures. Furthermore, the development of thermokarst lakes accelerates the decomposition of permafrost carbon stocks, releasing greenhouse gases such as CO₂, CH₄, and N₂O, which further feedback into the climate system. Currently, coupled water-heat-carbon cycling processes and their environmental implications represent a key research focus in permafrost science. Future studies should comprehensively consider the interactive effects of climate change and human activities and, based on coupled water-heat-carbon cycling processes, develop high-precision land surface process models to investigate ecological succession, water resource dynamics, and carbon cycling in permafrost regions under changing environmental conditions, thereby advancing cryospheric science.

Summer heat extremes are among the major meteorological disasters in China, posing severe threats to public health, economic and social development, and natural ecosystems. To address the nation's urgent need for managing heat-related disaster risks, we independently developed a prediction model system for summer heat extremes in China, based on new scientific insights. Since 2018, the model system has demonstrated stable and reliable predictive capabilities, relatively accurately capturing the spatial patterns and anomalies of summer heat extremes. In May 2025, using this system, we predicted that the number of summer hot days in 2024 would be 12.55 days, which is 2.69 days more than the average of normal years (1991-2020). The forecast also indicated more severe heat extremes, elevated disaster risks, and pronounced regional differences. The most significant above-normal heat extremes were expected in the middle and lower reaches of the Yangtze River Basin, South China, the Sichuan Basin, southern Xinjiang, northern Jiangsu, and northern Anhui. These were followed by the Beijing-Tianjin Plain, Shandong, Henan, southern Shaanxi, parts of northeastern China, parts of Gansu, and northern Ningxia. Based on these findings, we also provide response recommendations to prevent and mitigate the impacts of summer heat extremes across China.

Abnormal atmospheric warming on the Tibetan Plateau has caused an imbalance in Asian Water Towers, leading to widespread and frequent cryospheric disasters such as ice avalanches and Glacial Lake Outburst Floods (GLOFs). These events pose a significant threat to life and infrastructure downstream, impacting regional socioeconomic development. Our recent studies, conducted during the Second Tibetan Plateau Scientific Expedition and Research Program, utilized field observations, remote sensing, and modeling to examine glacial lakes and GLOFs on the Tibetan Plateau. As of 2020, we had identified 14 310 glacial lakes on the Tibetan Plateau, covering an area of 1 148.3 km², along with a 20.4% increase in lake number and a 20.2% increase in lake area since 1990. Hazard and risk assessments revealed 1 256 glacial lakes with high or very high hazard levels, including 182 glacial lakes with high or very high-risk levels. These high-risk glacial lakes pose severe GLOF threats to communities and infrastructure downstream. At the regional scale, the eastern Himalayan and southeastern Tibetan regions exhibit the highest number of glacial lakes, the largest area expansion, the most destructive GLOF hazards, and the highest concentration of very high hazard level and very high-risk level glacial lakes on the Tibetan Plateau. In terms of administrative regions, Shigatse City, Nyingchi City, and Shannan City in the Tibet Autonomous Region have the highest distribution of very high-risk level glacial lakes. Future research should focus on precise GLOF assessments, the development of monitoring and early warning systems, and strategies for adapting to GLOF disaster chains and transboundary threats.

Clearly defining mineral deformation and slip systems is crucial for an in-depth analysis of the intrinsic mechanisms governing mineral responses to external stress and temperature, as well as their rheological weakening processes. The rapid advancement of science and technology and its deep integration into the geological field provide an opportunity for a detailed analysis of structural deformation behavior and mechanisms. In this study, quartz and amphibole from representative naturally deformed rocks were used as examples. Based on microstructural analysis, a comprehensive assessment was conducted using a substantial dataset of mineral lattice preferred orientation measurements obtained via an electron backscatter diffraction (EBSD) probe mounted on a field-emission scanning electron microscope. By examining microstructural features, EBSD mapping data, dislocation geometry types, and properties, a detailed analytical method for grain boundary trace and misorientation axes was developed. The results reveal that the strain adjustment and grain refinement process in quartz occur mainly through the {m}<a> slip system, dominated by the subgrain rotational recrystallization mechanism in quartz veins. It was also found that in mylonitic amphibolites, amphibole porphyroclasts exhibit pronounced fine-grained deformation behavior, primarily driven by subgrain rotational recrystallization. Furthermore, amphibole undergoes multi-slip system interactions, predominantly governed by the [001] direction through dislocation creep in banded amphibolites. Thus, integrating EBSD grain boundary trace analysis with misorientation axis analysis and microstructural characterization enables a comprehensive determination of microgeological information—including composition, shape, grain size, orientation, boundaries, and strain—of deformed minerals. This approach further elucidates the evolution of orientation from the grain interior to intergranular regions (or matrix). Moreover, the dominant slip system in mineral deformation processes can be effectively defined and correlated with the deformation environment, which has substantial geological implications.

As the largest desert in the world, the Sahara Desert emits dust aerosols, accounting for 50%~60% of the global total dust, exerting significant impacts on regional and even global climate, environment, and ecosystems. Previous domestic and international studies reported two primary transport pathways for Saharan dust: westward across the North Atlantic, reaching North America, or northward to the European continent. In recent years, studies have shown that Saharan dust can be transported across the Middle East and Central Asia, undergoing long-distance (nearly 10 000 km) to East Asia, which is the third transport pathway for Saharan dust. Therefore, this study primarily summarizes the research progress on the long-range transport of Saharan dust to East Asia and its impacts, including the physical and chemical properties of Saharan dust, dust emission mechanisms, transport processes, and climatic and environmental effects. Finally, we highlight the current challenges in the research on the eastward transport of Saharan dust and provide suggestions and ideas for future research.

Continuous vertical gradient observations of aerosols, clouds and precipitation in mountainous terrain provide critical insights into their distribution characteristics in vertical direction. Mountain cloud observation is thus an effective way for studying the formation mechanism of cloud and precipitation. This article reviews the development and current status of mountain clouds and fog observation technology over the past century, and summarizes the domestic and international research results of mountain clouds and fog observation. Cloud droplet sampling technology has evolved through three main stages: collision sampling, laser scattering, and cloud particle imaging. Currently, laser scattering technology is the primary method for cloud particle measurement due to its reliability, meanwhile cloud particle holographic imaging technology has advanced significantly owing to its capacity to preserve particles’ natural morphology and ambient conditions. Europe pioneered in meteorological observations in mountainous regions. In China, the mountain clouds and fog observation began in the 1950s, promoting the physical study of clouds and precipitation. Over the decades, mountain cloud physical observation stations in China have covered several typical climate zones. The physical characteristics of aerosols, cloud condensation nucleus, atmospheric ice nucleating particles and clouds were obtained, meanwhile the formation mechanisms for warm and mixed-phase clouds were investigated. A comparison of global mountainous observations reveals that the cloud droplet number concentration typically ranges from 106 to 500 cm^-3, and the liquid water content typically ranges from 0.01 to 0.3 g/m³. Both parameters exhibit slight increases with altitude. Despite similar observation heights, the mean cloud droplet number concentration and liquid water content observed on Mount Lu were 45 cm^-3 and 0.05 g/m³ lower, respectively, than those observed on the Western Ghats of India, primarily due to monsoonal differences. Thus, cloud microphysical parameters are influenced by both regional climate and observation altitude. In terms of cloud formation mechanism, the research in European and American has focused on the characteristics of atmospheric ice nucleating particles and formation mechanism of mixed phase cloud due to high-altitude of observation stations, highlighting the important effects of factors (supercooled droplets and updrafts) on rimming and Bergeron processes as well as blowing snow mechanism. Warm clouds observation in mountainous areas have confirmed key processes, including collision, turbulence, and entrainment, as well as the vital roles of mid-sized cloud droplets and secondary peaks on precipitation formation. Frequent drizzle in mountain regions is closely associated with cloud condensation nucleus, ice nucleus, weak updraft, turbulence and high-humidity conditions. In contrast, China’s cloud physics observation stations are located near the top of the boundary layer and below the boundary layer, so early mountain cloud and fog observations in China captured fluctuations in warm cloud microphysics and environmental parameters, which accelerated the collision-coalescence processes critical for the generation of mid-sized cloud droplets. So early mountain cloud observations have significantly contributed to advancing warm cloud fluctuations theories. Finally, prospects and suggestions are proposed for the cloud observation technologies and studies in mountainous regions.

Significant advances in the formation mechanism and forecasting methods of severe convective winds and related convective systems were reviewed to improve understanding of the formation mechanism and forecast accuracy of severe convective winds. First, the spatial and temporal distribution characteristics of severe convective winds worldwide are briefly described. Next, the relationship between the organizational types and structural features of the parent convective systems that generate severe convective winds is then summarized, as well as the impact of atmospheric environmental conditions and topography, and forecasting methods. Finally, the current issues and future research directions associated with severe convective winds are discussed.

Northwest China is one of the world’s typical arid regions, where limited water resources severely constrain social development. However, the current utilization of atmospheric cloud water resources in this region remains significantly underdeveloped. Investigating the spatiotemporal variations of cloud water resources and cloud-precipitation processes is of great practical importance for enhancing technological capacity to exploit atmospheric water resources. To address this challenge, the National Natural Science Foundation of China (NSFC), through its Regional Innovation and Development Joint Fund, has supported the project “Multi-scale Variations of Atmospheric Cloud Water Resources and Cloud-Precipitation Processes in Northwest China”. This study highlights the strategic importance of developing cloud water resources in the region and examines the complexity of water formation and precipitation conversion mechanisms. Key influencing factors include the interaction of multiple atmospheric circulation systems; the macro- and microphysical complexities of cloud processes; the unique activation effects of dust aerosols; the topographic influences of plateaus and major mountain ranges; and the impact of regional climate warming and humidification. The critical role of field observations in supporting these investigations is also emphasized. Based on these insights, the study identifies six key research priorities for the future, including understanding variability patterns, aerosol-cloud interactions, cloud-precipitation conversion mechanisms, and advancing cloud microphysical parameterizations. These efforts aim to establish a robust theoretical and technical foundation for the effective utilization of atmospheric water resources in Northwest China.

As the climate crisis intensifies, Earth system models have become increasingly significant as critical numerical simulation tools for evaluating and addressing future climate change. The Coupled Model Intercomparison Project (CMIP), aimed at promoting model development and deepening the scientific understanding of the Earth's climate system, has become a central platform for international model exchange and application. This paper provides an overview of China’s participation in the Sixth Phase of CMIP (CMIP6), including a statistical analysis of citations, research trends, and key characteristics of the Chinese Earth system models in CMIP6-related studies. In addition, the Seventh Coupled Model Intercomparison Project (CMIP7), which is currently under preparation, is briefly introduced, and the opportunities and challenges faced by China in model development are summarized. Through continuous technological innovation, international cooperation, and exchanges, Chinese scientists are expected to make greater breakthroughs in the field of Earth and Climate System Models and contribute to Chinese wisdom and solutions for global climate change response and governance.

Carbon neutrality is a crucial strategy for combating global warming, and negative emissions technologies are key to achieving this goal. As the largest carbon reservoir on Earth, the ocean plays an irreplaceable role in regulating global carbon cycling and holds significant potential for negative emissions. Ocean alkalinity enhancement is a highly efficient and ecologically beneficial negative emissions technology. This technology increases ocean alkalinity by adding alkaline minerals to seawater, thereby enhancing the absorption of atmospheric CO₂ and improving the buffer capacity to resist ocean acidification. This study introduces the mechanisms and advancements in ocean alkalinity enhancement research at multiple scales based on the dissolution theory of carbonates in the ocean. Assessing the potential for negative emissions and associated costs reveals several challenges regarding implementation pathways, environmental impacts, and public acceptance. Considering the specific conditions of China’s coastal regions and the characteristics of ocean alkalinity enhancement technology, this study proposes a pathway integrated with wastewater treatment plants and coastal engineering. Furthermore, it provides an innovative concept on the application of ocean alkalinity enhancement and enriches the scientific understanding of blue carbon sinks.

The Huangshui River Basin, a region highly sensitive to climate change, was selected as a case study to investigate the evolution of extreme runoff at a regional scale and its climatic driving mechanisms. Daily average flow data were collected from seven stations in the basin. Mann–Kendall trend analysis and mutation tests were applied to assess the interannual variation of extreme runoff and its associations with extreme precipitation and high temperatures. The results indicate that over the past 60 years, the extremely high flow index in the basin has significantly decreased, whereas the extremely low flow index has notably increased. The frequency index did not exhibit any significant trend; however, all indices demonstrated persistence. Mutations in the high-flow index occurred around 2000, whereas mutations in the low-flow and frequency indices occurred in 2010. Cyclic analysis revealed that all indices exhibited a short cycle of approximately 3 years, whereas the frequency index also showed a long cycle of 32.5 years. Runoff variations were significantly correlated with an overall increase in extreme precipitation intensity, a decrease in precipitation duration, and an intensification of extreme high temperatures in the basin. Extremely high flows showed a positive correlation with extreme precipitation and negative correlation with extreme high temperatures. By contrast, extremely low flows exhibited a primary positive correlation with extreme high temperatures and weaker correlation with extreme precipitation. These findings provide critical insights for water resource management and flood disaster mitigation in the Huangshui River Basin.

Concerns about aviation emissions and climate change are shared internationally. The aviation industry plays a role in climate warming through its greenhouse gas and high-altitude particulate emissions. Conversely, climate warming alters flight conditions and increases extreme weather, impacts aviation operations and safety. The interaction creates a complex cycle of impacts, and research in this area is not only crucial for coordinating and adapting to climate changes in the aviation industry, but also holds scientific significance. An extensive literature review explores the relationship between aviation and climate warming, examining aviation’s CO₂ and non-CO₂ contributions to global warming and the phenomena and mechanisms by which climate warming in turn affects aviation (including changes in turbulence, flight time, aircraft performance degradation, and increased frequency of extreme events). The review also presents future research prospects. A deeper understanding of this interrelationship will help promote sustainable development of aviation and provide a scientific basis for addressing global climate challenges.

Human-earth system science, as a foundation of sustainable development research, can help decision-makers design sustainable pathways through multidimensional perspectives, integrated concepts, and systematic thinking. It plays an increasingly important role in the construction of national economies, societies, and ecological civilizations. Human-earth system sustainable development assessment models and scenario analysis techniques have become important tools that are widely used and studied. However, current research lacks a summary of the progress and limitations of these models and scenario analysis techniques. To keep pace with international developments and promote the understanding and advancement of human-earth system modeling and decision analysis of Chinese scholars, it is necessary to review the current international research in this field systematically. By combining literature analysis and quantitative analysis, this study summarizes the difficulty of models in simultaneously supporting multiple sustainable development goals and the challenges in simulating the social dimension. We also analyze the challenges in capturing systematic change, scale conversion, interdisciplinary knowledge integration, uncertainty management, data mining, and the use of new technologies. Additionally, we summarize the methods for setting up scenarios, the general types of scenarios, the content of scenarios, the limitations in addressing internal scenario conflict cross-scale linkages, and connections with decision-making. This study provides an important reference for promoting innovative development among Chinese scholars in this field.

The monitoring and early warning of pathogenic microorganisms and infectious diseases serve as a critical foundation for preventing major public health crises and mitigating biosecurity risks. However, research on the monitoring and early warning of pathogenic microorganism transmission in the atmosphere remains limited, with no systematic framework established yet. This study addresses strategic needs in public health security by identifying key scientific challenges in the field, systematically elucidating the environmental response mechanisms of atmospheric pathogens under climate change, monitoring technologies for pathogenic microorganisms in the atmosphere, and advances in infectious disease prediction models. Furthermore, this study identifies critical research frontiers for future breakthroughs, including: elucidating the source characteristics, formation mechanisms, environmental evolution, and transmission mechanisms of atmospheric pathogens; developing high-precision real-time monitoring technologies for atmospheric pathogens and establishing a biosafety surveillance network; constructing a multi-disciplinary, multi-scale and multi-model coupled prediction and early warning platform for atmospheric pathogen and infectious diseases. This research framework will provide scientific decision-making support for preventing public health emergencies, effectively enhance biosecurity governance capacity, and offer a scientific paradigm for building a global community of health for all.

Borehole collapse pressure prediction plays a key role in drilling safety, reducing construction costs, and realizing efficient drilling. Fracture development under complex ultra-deep geological conditions significantly affects the prediction of borehole collapse pressure. Conventional methods rely on finite element simulations for 3D geomechanical modeling and 3D collapse stress prediction, which although, highly accurate, requires substantial computational resources. To address this issue, the study proposes an efficient and rapid in situ stress modeling method driven by seismic data, utilized for 3D collapse pressure prediction. Initially, a combined spring model with curvature properties is developed using a multi-scale data of pre-stack seismic and rock mechanics logging to model a three three-dimensional in situ stress field efficiently and rapidly. Next, based on the maximum likelihood attribute, the fracture development was obtained from 3D seismic data to provide 3D weak surface attribute parameters for the study area. Finally, the collapse model of sliding along the fracture plane was calculated using the Mohr-Coulomb criterion. This enables the collapse pressure prediction of the fractured formation from one-dimensional logging data to a three-dimensional working area. This method was applied to the woodworking area of Tari, with results showing a high agreement between model predictions measured data, reaching 93.79%. The prediction results also aligned well with formation micro-resistivity scanning imaging interpretations, verifying the method’s feasibility for predicting borehole wall collapse events. This study demonstrates that rapid, high precision modeling of collapse pressure can provide an integrated geological engineering solution for drilling in ultra-deep and complex areas.

Debris flow disasters, known for their frequent occurrence and high destructiveness, are difficult to monitor effectively due to the limited real-time performance and high false-alarm rates of conventional monitoring methods. This critical limitation underscores the urgent need to develop highly efficient and precise intelligent detection techniques to substantially enhance early warning capabilities. To address the challenges of poor real-time performance and high false alarm rates in traditional debris flow monitoring systems, this study proposes an enhanced YOLOv8m-GCSlide model based on the YOLOv8 framework. The GlobalContext Network (GCNet) is integrated into the backbone network to improve spatial dependency modeling of dynamic fluid boundaries in complex terrains, while a Sliding Loss function (SlideLoss) is designed to dynamically adjust classification thresholds and mitigate sample imbalance. Knowledge distillation is applied to compress the model, resulting in a lightweight variant (YOLOv8n-GCSlide) with reduced computational complexity. A multi-source video dataset was constructed using publicly available resources, with frames extracted at 0.25-second intervals to balance feature retention and training efficiency. Data augmentation techniques, including random cropping, rotation, scaling, Gaussian blur, and color jittering, were used to enhance generalization, supplemented with negative samples (e.g., dry riverbeds and landslides) to reduce false positives. Experimental results show that the optimized model achieves 94.6% (+2.0%) detection accuracy, 88.0% recall, 95.9% mean Average Precision (mAP), and an inference speed of 244.1 FPS, outperforming mainstream lightweight models such as SwinTransformer and MobileNet variants. After compression, the model parameters were reduced by 88.1%, with the distilled version retaining 94.6% (+1.2%) accuracy and 88.0% (+0.7%) recall while maintaining an inference speed of 244.1 FPS. Field validation conducted in Sedongpu Gully, a high-risk debris flow region, confirmed the model’s practical applicability. Under complex environmental interference, the model achieved 82.3% recall, 4.2% false positive rate, and a processing speed of 240.6 FPS. The integration of global attention mechanisms and task-specific loss functions effectively captures dynamic motion features and suppress environmental noise. Additionally, model compression techniques help balance accuracy and computational efficiency, enabling edge deployment for real-time disaster warnings. This approach provides a robust technical foundation for intelligent geological hazard monitoring systems, emphasizing high precision, low latency, and adaptability to resource-constrained scenarios.

Wind and solar energy have unparalleled advantages in reducing greenhouse gas emissions and promoting energy transitions. However, the construction of onshore wind/solar farms occupies a tremendous amount of land resources and changes land use considerably. The operation of power generation facilities further changes the local microclimate and ecohydrological processes, profoundly affecting terrestrial carbon cycle processes. Therefore, it is important to clarify the potential impacts of wind/solar farms on the carbon cycle process at the site for sustainable development of the new energy industry. A systematic review of the research undertaken over the past two decades was conducted in this study, with special emphasis on the carbon cycle characteristics, impact mechanisms, and the dynamics and stability of carbon pools in onshore wind/solar farms. The results indicate that these wind/solar farms have the potential to improve local climate conditions, promote the restoration of vegetation, and thus increase the carbon sequestration potential in arid desert environments. However, considerable uncertainties exist regarding the recovery potential of either vegetation or soil carbon pools for wind/solar farms. We argue that there is an urgent need to \mathrm{①} conduct multi-scale and long-term monitoring of the carbon cycling processes in wind/solar farms, \mathrm{②} strengthen research on the synergistic mechanisms of the above- and below-ground carbon processes in onshore wind/solar farms, and \mathrm{③} quantitatively determine the carbon sequestration potential and its spatial and temporal characteristics in wind/solar farms. These efforts are expected to provide scientific references for sustainable design, management, and development of renewable energy sources in the future.

Since 1958, China has conducted numerous artificial fog dissipation field experiments and research. This paper summarizes the classification and characteristics of fog as well as the mechanisms and methods of artificial fog dissipation. Fog areas in China are extensively distributed, with obvious seasonal differences. Land fog is mostly radiation fog, whereas sea fog is distributed in foggy areas along the coast, and its formation and dissipation are restricted by various conditions. The methods and technical approaches for artificial warm and cold fog dissipation were determined. The dissipation methods for warm fog include heating, dynamic mixing, thermodynamic methods, and hygroscopic particle seeding; whereas the dissipation methods for cold fog include seeding silver iodide of ice nucleating agents and spraying refrigerants. Other methods such as ultrasound are currently being researched and tested. The applicability, advantages, disadvantages, and uncertainties of these seeding methods were analyzed. The applicability of the fog dissipation methods varies. When applying these methods, it is necessary to comprehensively consider the technical approaches, implementation challenges, cost-effectiveness, and fog dissipation efficacy in field trials and operational applications. Aircraft-induced downdraft mixing is a simple, expensive, and operationally challenging process for warm fog. Thermal heating is universally applicable to all warm fog types but is cost-prohibitive and reserved for emergencies or critical infrastructure (e.g., major international airports and vital seaports), particularly for high-temperature fog. For cold fog, silver iodide seeding exhibits poor nucleation efficiency at temperatures around -5 ℃ (optimal below -8 ℃), necessitating cooling agents like liquid nitrogen, dry ice, and propane. Despite its high cost, liquid-nitrogen seeding is preferred operationally owing to its reliability and ease of deployment.All the current methods can dissipate local small-range warm or cold fog, but none can dissipate large-scale fog systems. A comprehensive analysis of fog dissipation provided ideas and references for artificial fog dissipation experiments, seeding operations, and future development in China. Future research should integrate numerical modeling, laboratory experiments, and field trials to validate and optimize seeding techniques and enhance the operational efficiency and cost-effectiveness.

Submarine landslides are among the most common and destructive geological hazards on continental margins. Their development can significantly reshape seafloor morphology, generate high-density turbidity currents, and even trigger catastrophic tsunamis, posing serious threats to the safety and operation of sub-sea engineering infrastructures. The formation of submarine landslides typically involves long-term geological processes influenced by multiple interacting factors. Some large-scale submarine landslides exhibit multi-stage sliding events with complex movement histories. However, current understanding of the developmental characteristics and formation mechanisms of multi-phase submarine landslides remains limited, which hinders scientific insight into their evolutionary patterns. Based on high-resolution, two-dimensional (2D) multichannel seismic and borehole data, six phases of Mass Transport Deposits (MTDs) resulting from submarine landslides have been identified in the Kaiping Sag on the northwestern continental margin of the South China Sea. According to the established regional sequence stratigraphic framework, these MTDs are mainly concentrated within the Lower Hanjiang Formation and Yuehai Formation. Seismic interpretation results indicate that the internal structure of MTD 1 and 2 is highly deformed and significantly altered by subsequent geological processes, whereas MTD 3 to MTD 6 exhibit typical landslide features such as prominent headwall scarps and lateral margins. Calculation of sedimentation rates during the occurrence of each phase of MTDs reveals that high sedimentation rates occurring during periods of low sea level provided the necessary sediments for the occurrence of landslides. This rapid sediment accumulation likely prevented the timely expulsion of pore fluids, leading to elevated pore pressure within the sediments and the formation of unstable weak layers. In addition, the widespread development of tectonic normal faults (e.g., Shenkai Fault) and their intersecting relationships with all six MTDs strongly suggest that fault activity also played a significant role in triggering these landslides. This study provides new insights into the formation mechanisms of submarine landslides along the northwestern continental margin of the South China Sea, offering important scientific support for hydrocarbon exploration, geological hazard risk assessment, and disaster prevention and mitigation in the region.

Short-duration heavy precipitation is one of the most substantial severe convective disasters in China and is prone to causing urban waterlogging and secondary geological disasters, such as mountain torrents, mudslides, and landslides. This paper reviews recent progress in short-duration heavy precipitation research in China and briefly compares relevant findings from the United States and Europe. It covers the spatiotemporal distribution characteristics and diurnal variation patterns of short-duration heavy precipitation, atmospheric circulation patterns and environmental conditions that influence its occurrence and development in major regions of China, radar echo characteristics and raindrop distributions, impact of topography and urbanization on its formation and development, and application of artificial intelligence in potential forecasting, short-term forecasting, and nowcasting of short-duration heavy precipitation in China. With global warming, the frequency and intensity of short-duration heavy precipitation events have increased. In the future, further research will be required to enhance understanding of the formation mechanisms and environmental conditions, improve the spatiotemporal resolution of observations, expand the use of new observation data, and enhance forecasting capabilities in high-resolution, rapid-update cycle assimilation numerical weather prediction models through the fusion and analysis of dense multisource observation data. Additionally, optimizing deep learning models and algorithms—particularly in the development of largescale deep learning models—will be crucial for improving forecasting and early warning capabilities for short-duration heavy precipitation.

Humanity’s current water problems range from local-scale issues such as water supply to regional- and global-scale issues including protecting ecosystems, responding to global changes, sustaining the earth system, etc. Water resources exploitation, land utilization and climate changes have intensified pressure on water cycle through water distribution, interconnection, and virtual flows. The impact of anthropogenic pressure on water cycle has extended beyond the catchment-scale, with human activities becoming the primary driving force behind changes in regional, continental and global water cycle. Estimations by planetary boundaries framework indicated that development of global blue water and green water is approaching or beyond water planetary boundaries posing increased rising risks to earth system stabilization. Current water governance, which is focused on catchment scale and water-centric approaches, struggles to address the complexity of these issues. Governance must shift to manage not only increasing water use for economic and societal development, but also the roles and functions of water cycle in sustaining biosphere and Earth systems. Moreover, it should consider the equitable distribution of ecological services provided by water cycle. Concepts of water resilience and the economics of water as a common good enhance the conventional understanding of the water cycle, highlighting its essential role in sustain Earth systems and the cross-scale effects of human activities. Future, water resources governance is likely to evolve in three directions: from blue water management to blue-green water management, from integrated water-centric management to integrated land-water-ecosystem management, and from integrated river basin management to multi-scale management. It is critical for promoting transformation of water governance to strengthen cooperation among scientists of different fields in research of basic theory of water cycle, management policies and governance institutions.

Vegetation fire, as a significant disturbance factor in the Earth’s system, can have important impacts on Earth’s surface systems, such as the atmosphere, hydrosphere, cryosphere, biosphere, pedosphere, and anthroposphere. The various gases and aerosol particles released during vegetation fires not only affect the atmospheric environment, but also pose risks to human health. Extreme vegetation fires can cause serious casualties and economic loss. In recent years, under the influence of global warming, various extreme weather events have occurred frequently, and the risk of vegetation fire disasters has also significantly increased. Understanding the stages and mechanisms of modern vegetation fires is of great scientific significance for predicting future changes in vegetation fires, and is of great practical importance for formulating fire management strategies. This study provides an overview of the research progress on vegetation fires in China from the perspective of the evolutionary history of vegetation fires and modern vegetation fire regimes. The following basic understanding is obtained regarding the pattern of modern vegetation fires: First, from historical records, modern vegetation fires are currently at their most frequent period since the Holocene, and from the late 20th century to the present, even to the mid-21st century, vegetation fires in China show an overall upward trend. Second, vegetation fires in China are mainly agricultural fires with forest wildfires as a supplement, concentrated in spring and autumn, and are mainly distributed in northeastern, southwestern, eastern, and southern China, showing regional diversification characteristics under the influence of human activities and climate change. In the future, efforts should be made to strengthen the review of the details of historical changes in vegetation fires in China, elucidate the overall modern vegetation fire regime, and provide more accurate predictions of future changes in vegetation fires.

Qilian Shan, the youngest mountain range formed by the northward expansion of the Tibetan Plateau, plays a crucial role in understanding the expansion processes, uplift mechanisms, and evolution of orogenic belts. Drainage system evolution responds rapidly to mountain uplift, making the study of drainage development and evolution a critical approach for investigating the uplift and expansion of Qilian Shan. Based on chronological and provenance studies of geomorphic records, including erosion surfaces, river terraces, wind gaps and ancient river channels, and Cenozoic sedimentary strata, the current research on drainage system evolution in the Qilian Shan has yielded the following findings and insights: \mathrm{①} The formation and evolution of the upper reaches of the Yellow River in the eastern Qilian Shan involve a process of drainage reorganization driven by tectonic uplift or climate change, characterized by headward erosion and river capture; \mathrm{②} Research on river terraces in the Shiyang River and Heihe River basins of the northern Qilian Shan, as well as in the Lanzhou Basin of the eastern Qilian Shan, indicates climate change, and the tectonic uplift independently govern the timing (transitions between glacial and interglacial periods, and interglacial periods) and extent of river incision. Since the Holocene, terrace formation has been primarily driven by climate change, with river incision occurred during warm and humid periods; \mathrm{③} River terraces reliably record the evolution processes of major tributaries of the Yellow River in the eastern Qilian Shan, including the Huangshui River (flow reversal) and the Datong River (river capture); \mathrm{④} Study of chronology, provenance, and paleohydrology of Cenozoic sedimentary strata in the Yumu Shan of the northern Qilian Shan, as well as the Wulan and Chacha basins of the southern Qilian Shan, has reliably reconstructed the regional drainage evolution history, highlighting the significant potential of sedimentary strata for reconstructing reliable and detailed record of drainage evolution. However, numerous critical issues remain unresolved and require further investigations. Future research should prioritize and emphasize in-depth studies on geomorphic surface and sediment dating, integration of multi-source methods for provenance analysis, continuous exploration of geomorphic features, and advancements in numerical simulations and simulation modeling studies.

Subsea permafrost, formed by the inundation of terrestrial permafrost due to sea-level variations during the interglacial cycles, is primarily distributed across the Arctic continental shelves. However, a substantial uncertainty remains regarding the extent of its distribution (approximately 1~2.7 million square kilometers). Subsea permafrost is considered a significant carbon reservoir in the Earth’s system, storing vast amounts of Organic Carbon (OC) and methane (CH₄). With global warming and rising Arctic Ocean temperatures, subsea permafrost is undergoing rapid degradation, potentially exacerbating carbon release risks. Consequently, it plays a significant role in the global carbon cycle and climate dynamics. Large-scale CH₄ emissions into the atmosphere have been observed in the East Siberian subsea permafrost region. However, the rates of subsea permafrost degradation, the size of carbon reservoirs, and gas release remain poorly constrained. In particular, rapid Arctic warming, the northward expansion and intensification of the North Atlantic Current (which exacerbates the Atlantification of the Arctic Ocean), and increased human disturbances have intensified climate risks due to accelerated CH₄ emissions from Arctic subsea permafrost. These changes have significant implications for future human sustainability. This study systematically summarizes the spatial distribution, degradation rates, and carbon storage of Arctic subsea permafrost. It also examines CH₄ monitoring in subsea permafrost, including fixed-point observations, aerial surveys, and remote sensing technologies. Furthermore, it discusses the factors influencing CH₄ emissions, emphasizes the importance of understanding Arctic subsea permafrost dynamics within the context of global climate change, identifies key challenges, and suggests future research directions.

Evaluating the distribution and sea-air fluxes of dissolved methane (CH₄) in mariculture areas is important for understanding how aquaculture contributes to regional CH₄ emissions into the atmosphere. Seasonal field surveys conducted in 2023 were used to analyze the temporal and spatial variation of CH₄ concentrations in surface water and CH₄ air-sea flux in a typical aquaculture system in Sansha Bay, Fujian Province. The results showed that dissolved CH₄ concentrations ranged from 9.91 to 609.22 nmol/L, with corresponding air-sea fluxes between 3.46 to 1 188.15 μmol/ (m²·d). Temporally, the CH₄ air-sea fluxes were higher in summer and autumn compared to spring and winter. Spatially, CH₄ concentrations and air-sea fluxes decreased consistently from the estuary to the bay mouth, with the highest values in the estuarine aquaculture area and the lowest in the bay mouth aquaculture area. Correlation analysis showed that aquacultural activities and terrestrial runoff inputs contributed to the spatiotemporal distribution of CH₄ concentrations within the bay. In the macroalgae cultivation zones, CH₄ production and emissions during farming periods were significantly lower than during non-farming periods. Additionally, the residual feed and feces generated by fish in cages may result in increased CH₄ emissions. Notably, CH₄ emissions peaked in summer, due to enhanced aquaculture activities and runoff inputs during the wet season. Future work should focus on investigating CH₄ air-sea fluxes in mariculture areas to provide scientific support for CH₄ control and emission reduction in aquaculture.

Ngari is a unique desert steppe type on the Tibetan Plateau, characterized by its desert steppe ecosystem. This region serves as an ideal representation for studying the quantitative relationships between vegetation and climate. Relative Pollen Productivity (RPP) and Relevant Source Area of Pollen (RSAP) are important parameters for quantitative studies of vegetation and climate based on pollen. Based on modern topsoil pollen and vegetation data from 37 sample sites in the Ngari Desert steppe, different sub-models of the ERV model were utilized, with Chenopodiaceae pollen as the reference species, to estimate the relative pollen production and relevant source area of pollen for five major pollen types: Poaceae, Chenopodiaceae, Artemisia, Asteraceae, Brassicaceae, and Potentilla. The results showed that sub-model 2 provided optimal estimates. The relevant source of pollen in the study area was 1 550 m. The relative pollen productivities of the main pollen types are as follows：Chenopodiaceae (=1.000), Artemisia (1.286±0.058), Asteraceae (0.689±0.043), Brassicaceae (0.763±0.063), Potentilla (0.139±0.008), and Poaceae (0.003±0.006). The results of the leave-one-out method and the REVEALS model validation indicate that the above RPP and RSAP results are reliable and can be applied to regional vegetation reconstruction.

Tracking the Earth’s energy imbalance is one of the key methods for studying the contribution of human activities to climate change. Energy imbalance directly reflects the complex responses and feedback of the climate system and is an important indicator of climate change. However, accurately estimating the Earth’s energy budget has long been a challenge. Observations of the top of the atmosphere and surface radiative fluxes have high uncertainties, and it is difficult to validate different observation datasets. In addition, these high uncertainties lead to inaccurate estimates of changes in Earth’s energy budget fluxes. Furthermore, estimating surface radiative fluxes is challenging because of the lack of high-quality, high-resolution observational data. Recently, methods using ocean heat content/sea-level height data for the indirect estimation of the Earth’s energy budget have been widely applied. Considering that most of the energy imbalance flows into the ocean heat content, ocean data observations can yield estimates of the Earth’s energy imbalance with lower uncertainty. Additionally, reasonable estimates of the Earth’s energy budget can be obtained through multi-model ensemble methods using Earth system model outputs, supplemented with appropriate weighting strategies. By improving data integration capabilities and developing related technologies, climate scientists continuously enhance their understanding of the Earth’s energy budget, providing more precise scientific evidence for understanding and addressing increasingly severe global warming.

Passive Microwave Brightness Temperature (PMWBT) is crucial for retrieving various land surface parameters. However, PMWBT images often exhibit many missing observations, particularly in low-latitude areas, owing to the limited coverage of polar-orbit satellites equipped with PMW radiometer imagers. Filling these gaps is essential to enhance the spatiotemporal integrity and application potential of PMW-derived products. To better understand the problem and propose solutions, the PMW radiative transfer theory and the reasons for the observation gaps are comprehensively reviewed. Subsequently, two filling approaches, multi-source data filling and effective data reconstruction, which are commonly used in remote sensing, were introduced and assessed for their suitability in filling PMWBT gaps. Upon reviewing related research and existing issues in filling PMW BT orbital gaps, it was observed that current studies on filling satellite-borne passive microwave brightness temperature orbital gaps are limited, and all use multi-source data with low generalizability because of sensor differences. In conclusion, the current research status and challenges are succinctly summarized. Furthermore, from the perspective of using reanalysis data and time-series modeling, the construction of a high-precision, general reconstruction method under special underlying surface conditions was explored.

Carbon and oxygen isotopes of benthic foraminifera are widely used for paleoenvironmental reconstructions. However, large benthic foraminifera (LBF) shells, as the predominant sediment type in coral reef areas, exhibit isotope values influenced by many factors, especially the “vital effect”, which limits their application. Therefore, this study systematically categorizes the main factors contributing to deviations in the carbon and oxygen isotope values of LBF, including symbiotic algae, the calcification process, individual development, and seasonal variation. Furthermore, the mechanisms underlying these factors are thoroughly examined. Additionally, the potential applications of LBF carbon and oxygen isotope indices are analyzed. Despite the influence of vital effects, these indicators can still serve as powerful tools for paleoenvironmental reconstruction in coral reef areas by selecting suitable species, employing micro-area analysis, and integrating these indices with other paleoenvironmental proxies.

Soil improvement plays a critical role in enhancing the soil carbon sink capacity and optimizing karst processes under global climate change. Using bibliometric methods, this study analyzed literature from Chinese and English journals (1990-2024), focusing on the regulatory mechanisms linking soil improvement to soil carbon cycling and karst carbon sinks. The Web of Science (WoS) and China National Knowledge Infrastructure (CNKI) databases indexed 712 and 468 relevant articles, respectively, most of them original research papers. Article numbers showing a fluctuating upward trend over the 34 years. This process can be divided into three phases: germination (before 2005), growth phase (2005-2013), and rapid development (after 2013). English-language journals published more articles than Chinese-language journals. The English-language literature initially focused on biochar carbon sequestration mechanisms and greenhouse gas emissions. Over time, the focus shifted towards the synergistic regulation of soil microbial functional genes and nitrogen-phosphorus nutrient cycling, reflecting a transition from mechanism-based analysis to application-oriented microbial-nutrient coupling. Chinese-language literature has expanded from monitoring basic indicators, such as soil respiration, moisture, and heavy metals, to systematic research on soil aggregate regulation, microbial community optimization, and improvement techniques. In addition, a Carbon-Pool Management Index (CPMI) is established to guide practical applications. In low-productivity karst regions, soil improvement efforts primarily assesses improvement measures on karst processes and the soil carbon balance. This study explored the bidirectional coupling relationship between karst carbon sinks and soil improvement. Specifically, high concentrations of soil CO₂ drive carbonate rock weathering, whereas improvement measures enhance the carbon sink effect by improving soil quality. To advance soil improvement research in karst areas, it is recommended to establish a quantification method for measuring the increase in karst carbon sinks due to soil improvement and develop a database. Furthermore, considering the calcium-rich and alkaline characteristics of karst soils, the synergistic effects of improvement measures on ecological restoration and sustainable agricultural should be evaluated to provide scientific support for global carbon neutrality. By leveraging soil improvement technologies, we can enhance soil and karst carbon sinks, address climate change more effectively, promote the integration of ecological restoration and sustainable agricultural development in karst regions, and contribute to the achievement of China’s Dual Carbon Goals.

Gravity Waves (GWs) significantly influence structure of the entire atmosphere and coupling between atmospheric layers. Research on gravity waves is crucial for deepening our understanding of atmospheric dynamics and for improving the accuracy of atmospheric models. While gravity waves are well-known in the fields of astronomy and physics, they also play a vital role in atmospheric science, particularly in the study of airflow, wave propagation, and climate variability. This review highlights the following key findings: ① Satellites are suitable for observing the middle and upper atmosphere; radar is most effective for detailed observations of vertical wave propagation; and reanalysis data are best suited for analyzing global GW characteristics; ② Compared with non-orographic gravity waves, orographic gravity waves generally have longer vertical wavelengths and can propagate to higher altitudes; ③ Orographic gravity waves are easier to trace due to their relatively fixed sources; and ④ Common parameterization schemes effectively simulate the drag effects of orographic gravity waves, while single-wave and global spectral techniques can predict the east-west momentum flux of non-orographic gravity waves. However, the complete generation and evolution processes of both types of GWs cannot yet be accurately simulated. There is still considerable room for improvement in the observation, identification, feature analysis, and parameterization of gravity waves. In the future, advancements in observational technology are expected to yield higher-quality data, enabling a clearer understanding of GW characteristics. Based on this, progress in parameterization methods and the application of artificial intelligence techniques is anticipated to enhance our understanding of the formation mechanisms of both orographic and non-orographic gravity waves, thereby improving the accuracy of weather and climate simulations.

Groundwater-dependent vegetation is essential in arid ecosystems, where it maintains ecological balance and supports biodiversity. The health and functionality of this vegetation are closely linked to groundwater characteristics, including groundwater quality, distribution, and fluctuations. This review explores the relationship between vegetation and groundwater, methods for identifying groundwater-dependent vegetation, the impact of groundwater on the plants, adaptation mechanisms of these plants, and the nonlinear dependencies and thresholds of vegetation in groundwater environments. The objectives of the study are to provide a theoretical foundation for protecting and restoring arid ecosystems and to provide support for the sustainable development and utilization of groundwater resources. Future research should focus on plant responses to groundwater changes at the individual, population, and community scales; the effects of climate change and human activities on groundwater-dependent vegetation; innovative methods for studying ecosystem resilience and state-transition mechanisms for groundwater-dependent vegetation; and identifying stable water environment factors and catastrophic thresholds for typical groundwater-dependent vegetation.

The thermal regime of soil is vital for determining the presence and thermal stability of permafrost. To study long-term changes in the permafrost thermal regime in the Headwater Area of the Yellow River (HAYR), a mathematical model for soil heat transfer to simulate the dynamics of ground temperatures at six boreholes using the HYDRUS-1D model. The model’s reliability and applicability were confirmed through parameter calibration. Changes in the permafrost thermal regime from 1979 to 2018 in the HAYR were then simulated using monthly air temperature data from the China Meteorological Forcing Dataset (CMFD). Model simulations revealed an abrupt change in the mean annual ground temperature in the HAYR after 1999. Prior to 1999, the changing rates were from -0.037 to 0.026 °C/a, whereas after 1999, they ranged from 0.0059 to 0.12 °C/a. The abrupt increase in mean annual air temperature in 1998 and the occurrence of extreme climate disasters in 1999 were identified as the primary reasons for the sudden changes in the permafrost thermal regime in 1999. The rise in permafrost temperature and decrease in its thermal stability are expected to impact water resource conservation and biogeochemical cycles. This study provides scientific and technological support for understanding the response patterns of plateau permafrost to climate change and strengthening the zoning and control of the ecological environment in the HAYR.