Since the operation of Three Gorges Reservoir, the Middle Yangtze River has entered a strong erosion process, leading to the drastic adjustment of the channel geometry, which may threaten the flood control safety and ecological health of the river channel. Therefore, it is crucial to clarify the magnitude and mode of the non-equilibrium evolution adjustments. This paper focus on the Middle Yangtze River, and simulates the spatiotemporal adjustment process following operation of the Three Gorges Reservoir by using the stablished spatiotemporal equilibrium response equation. The results indicate that: (i) In terms of adjustment potential, there are significant differences in each reach. Laterally, the channel width of YC-ZC Reach and Upper Jingjiang Reach remains unchanged, while the broadening potential in the Lower Jingjiang Reach and other lower reaches is slightly larger, showing a widening range of about 20~30m. Vertically, the adjustment potential of accumulated riverbed degradation in most reaches is close to or more than 2.0m, except for the meandering Lower Jingjiang Reach. (ii) In terms of adjustment mode, the YC-ZC Reach and Upper Jingjiang Reach controlled by bedrock nodes are dominated by vertical riverbed incision, while the Lower Jingjiang Reach, CLJ-HK Reach and HK-HK Reach affected by river bank collapse are dominated by slight broadening and vertical degradation. The above results can provide scientific basis for flood control planning and river regulation in the Middle Yangtze River.

After the impoundment and operation of the Three Gorges Project, the water and sediment conditions in the middle and lower reaches of the Yangtze River have changed significantly, and the impact of sandbar evolution on flood control safety has become increasingly prominent. Taking the Jingjiang reach in the middle Yangtze River as the study area, this research investigates the evolution of sandbars and its impact on flood control under the new water and sediment conditions. The results show that, following the impoundment of the Three Gorges Reservoir, the evolution of sandbars exhibits several notable features: marked differentiation in erosion patterns at different elevations; erosion processes concentrated in the flood season; the growth and decline of sandbars strongly influenced by the release of clear water; and predominantly erosional trends on point bars. The impacts on flood control are manifested as follows: in meandering reaches, point bars on convex banks are dissected by momentum exchange between the main channel and the bar and by secondary reciprocating flow (e.g., at Tiaoguan), while point bars on concave banks are scoured by centrifugal forces (e.g., at Qigongling). In braided reaches, the stability of bar bodies decreases and mid-channel bars shrink (e.g., at Taipingkou). In straight transitional reaches, oblique inflow scours the toes of sandbars (e.g., at Wuguizhou). The evolution of sandbars triggers shifting of the main flow, weakens the river-regime control function, and leads to changes in flow impingement and near-bank scouring, thereby endangering levee stability and threatening regional flood control safety. In response, countermeasures are proposed, including strengthening the monitoring and early warning of sandbar evolution, accelerating the systematic treatment of riverbank collapse, improving the emergency protection system, establishing a long-term mechanism for emergency protection, and deepening research on sandbar evolution. The findings can provide scientific support for flood control, disaster reduction, and river management.

Abstract In recent decades, global warming‐driven ocean stratification has profoundly affected nutrient cycling. As the largest marginal sea in the northwestern Pacific, the South China Sea (SCS) provides a representative setting to examine these responses. Based on two decades of temperature, salinity, and nutrient datasets, we quantified the evolution of stratification in the SCS, its impacts on the upward transport of deep‐water phosphate (PO 4 3− ), and the resulting changes in carbon sink. Results show an increasing intensification of stratification, with an upward shift of the thermocline's upper boundary, a downward shift of its lower boundary, increased thickness, and weakened vertical mixing. Despite enhanced external inputs from rivers and atmospheric deposition and reduced phytoplankton consumption of PO 4 3− in the mixed layer, PO 4 3− concentrations have declined persistently, indicating a substantial reduction in nutrient flux from the deep water across the thermocline into the surface layer. The minimum net reduction of PO 4 3− in the upper water column (mixed layer plus thermocline) was −0.047 Gmol·yr −1 . This decline in PO 4 3− supply constrained primary production and reduced CO 2 uptake, leading to an estimated carbon sink reduction of 2.21 × 10 5 t CO 2 ·yr −1 in the SCS.

Satellite remote sensing plays a fundamental role in observing oceanic processes by providing large-scale, long-term, and continuous measurements. With the increasing availability of multisource satellite data, challenges such as data gaps, complex environmental conditions, and the limitations of conventional retrieval methods have become more evident. In recent years, artificial intelligence (AI) has emerged as a practical and effective approach to address these issues. This article reviews the development of AI techniques in satellite ocean remote sensing, focusing on three main application areas: parameter retrieval, data reconstruction, and image-based ocean phenomenon detection. For geophysical variable retrieval, AI models such as convolutional neural networks (CNNs) and Transformer architectures have improved the accuracy of ocean waves, sea surface, salinity, wind, and ocean color estimates, especially under extreme or noisy conditions. In the field of data reconstruction, AI methods enable the completion of missing data in both surface and subsurface ocean layers, offering finer spatial-temporal resolution and better consistency than traditional interpolation approaches. For image interpretation, deep learning (DL) models have been applied to detect and segment dynamic ocean features such as mesoscale eddies, internal waves, sea ice, and tropical cyclones (TCs), achieving high efficiency and precision. This article also highlights the integration of AI with physical knowledge, the use of multisource fusion, and the trend toward near real-time (NRT) applications. These developments indicate that AI will play an increasingly important role in future satellite-based ocean observation and environmental monitoring.

Renewable-dominated Power System (RDPS) is emerging as critical trend to achieve CO2 emission reduction, and the COP28 climate summit has set a global goal to triple its capacity to at least 11,000 GW by 2030. However, this system is heavily reliant on weather and climate conditions. While the inclusion of the storage in the RDPS can alleviate the impact of the volatility and intermittency, the new source-grid-load-storage system creates more risks resulting from increasing complexity of interconnections. These include heightened variability in renewable energy generation under extreme weather, growing instability in transmission infrastructure due to climate-induced stress, performance degradation of storage systems under temperature extremes, and limited flexibility in demand response during climate-driven load surges. Critically, these risks are not isolated but often interdependent, as climate-triggered disruptions in one part of the system can cascade across interconnected components and amplify local failures into large-scale outages. In this perspective, we assess the risks for the system across source, grid, load, and storage from the dimensions of hazard, exposure and vulnerability, highlight the emerging cascading risks driven by climate variability and extremes, and call for an urgent expansion of climate risk assessment frameworks to ensure the resilience and reliability of future power systems.

The study of the water and sediment transport, river morphology adjustment mechanism caused by human activities, especially the construction of large reservoirs, is of great significance for maintaining the health of river system and realizing the sustainable development of river basin. After the impoundment of the Three Gorges Reservoir (TGR), due to a sharp decrease in the sediment load entering the middle reaches of the Yangtze River (MYR), the river channel downstream of the TGR have changed from the relative equilibrium of erosion and deposition to a long-term and long-distance erosion, leading to an important impact on the flood control and eco-environment safety. Therefore, it is urgent to carry out the research on the response adjustment process and the relative equilibrium state prediction of the riverbed morphology in the MYR. Based on the measured discharge, sediment load, water stage, and erosion and sedimentation data from the MYR, the response process of different channel morphological variables to the operation of the TGR is analyzed. Moreover, a generalized model for the quantitative simulation of its geomorphic adjustment is proposed by using the existing theoretical model of the accumulation processes of internal feedback induced by external disturbances. Results show that the response adjustment rates of the different channel morphological variables, including the cumulative erosion per unit river length, low-flow channel area and the fall values of water level at 10000 m3/s discharge, over time are relatively fast in the early period following disturbances but then slow down rapidly. The non-equilibrium adjustment process can be accurately described by this theoretical model, since the corresponding calculated values of the different morphological parameters were in good agreement with the measured values, the coefficient of determination (R2) between them remain at around 0.96. The attenuation coefficient of response intensity α of the model decreases downstream the channel, which is subject to the response law of the river morphology to the external disturbances that the response is earlier in the upstream than in the downstream and it continues to develop to the downstream.

Changes in nearshore water quality directly influence ecosystem stability and the sustainability of aquaculture production. Among these factors, rapid fluctuations in dissolved oxygen (DO) can compromise the physiological functions of aquatic organisms, often leading to mass mortality events and significant economic losses. To enhance the predictive capability of DO in marine ranching areas, this study evaluates multiple forecasting approaches, including AutoARIMA, XGBoost, BlockRNN-LSTM, BlockRNN-GRU, TCN, Transformer, and an ensemble model that integrates these methods. Using hourly DO observations from coastal buoys, we performed multi-step rolling forecasts and systematically assessed model performance across multiple evaluation metrics (MAPE, RMSE, and R2), complemented by residual and error distribution analyses. The results show that the ensemble model, based on deep learning techniques, consistently outperforms individual models, achieving higher forecast robustness and more effective variance control, with MAPE values maintained below 4% across all three buoys. Building upon these findings, we further developed and deployed a DO forecasting and early-warning system centered on the ensemble framework. This system enables end-to-end functionality, including automatic data acquisition, real-time prediction, hypoxia risk identification, and alert dissemination. It has already been applied in marine ranching operations, providing 1–3 day forecasts of DO dynamics, facilitating the early detection of hypoxia risks, and significantly improving the scientific support and responsiveness of aquaculture management.

China’s new urbanization and rapid social economic development have led to a significant increase in carbon emissions, while their precise monitoring plays a crucial role in carbon emission reduction and assessment of sustainable development goals (SDGs). For the shortcomings of the temporal lag and insufficient spatial scale of statistical data on CO2 emissions, as well as the low estimation accuracy, this study considers the differences in CO2 emissions among different industries and sectors, designs a spatiotemporal fine monitoring method for Beijing’s CO2 emissions from 2015 to 2020, combining with NPP-VIIRS NTL images, point of interesting (POI) data, grid population and residential land data. The distribution characteristics of energy consumption CO2 emissions in lit/unlit areas of 35 industries and residential areas, such as agriculture, forestry, mining and metal ore, are analyzed. The CO2 emissions in lit areas account for a larger proportion, up to 99%, while in unlit areas is relatively small at 1%. The density of population and residential areas in central urban areas will produce more CO2 emissions. This study can help to formulate appropriate emission reduction policies, optimize the energy structure, and build a green and low-CO2 economic development pattern.

Carbon emissions are currently a hot topic in the international community. CO2 reduction from heavy-duty commercial vehicles plays a significant role in slowing down the global greenhouse effect and promoting sustainable development. To control carbon emissions, many countries have tightened CO2 emission regulations and policy requirements for heavy-duty commercial vehicles in recent years. Various CO2 emission simulation models have been developed, such as the Greenhouse Gas Emissions Model (GEM) in the United States and the Vehicle Energy Consumption Calculation Tool (VECTO) in the European Union, to evaluate the real CO2 emission levels of commercial vehicles and provide a scientific basis for formulating corresponding emission reduction policies and control measures. This paper systematically analyzes the CO2 emission regulations and policy requirements for heavy-duty commercial vehicles in the United States, the European Union, China, and other developed countries. It also analyzes the GEM software in the United States, the VECTO software used in Europe, and the energy consumption simulation software for commercial vehicles in China. The influencing factors of CO2 emission simulation are explored in detail. This study found that, although GEM and VECTO software are recognized for their high accuracy, their applications are still dependent on local policies. In other countries and regions, VECTO software has broader applicability. On the other hand, China’s commercial vehicle energy consumption simulation software and other reported studies have only been validated for specific vehicle types. The accuracy and generalizability of these models should be further promoted and verified.

Due to global warming and the excessive input of nutrients resulting from human activities, ocean deoxygenation is gradually intensifying. However, the sparse spatiotemporal distribution of in situ global dissolved oxygen (DO) observations poses a significant challenge to understanding the spatiotemporal variations of DO in the world’s oceans. In this study, we employed a SOM-FFNN approach to reconstruct a global monthly dissolved oxygen dataset from 1960 to 2021, with high spatial (1° × 1°) and vertical (to 2000 m) resolution. Compared to existing products, our model exhibits a lower RMSE (15.36 μmol/kg), a higher R² (0.96), and strong agreement with long-term observations, outperforming other available datasets. Analysis of this dataset shows that the global ocean has been experiencing a steady loss of total oxygen content since the late 1980s, with the rate of depletion over the past decade nearly twice as high as that during 1980–2010. Further investigation reveals that low-oxygen zones have expanded both horizontally and vertically, with this expansion also intensifying in the past decade. These findings highlight the accelerating nature of ocean deoxygenation and the growing extent of low-oxygen habitats.

Water conveyance channels, as critical components of water diversion projects, feature numerous structures, complex configurations, and intensive operational management requirements, making them vulnerable to multiple risks, such as extreme flooding, channel blockage, structural failures, and management deficiencies. To ensure an accurate assessment of the operational safety risk, this study proposes a comprehensive risk assessment framework that integrates risk probability and risk loss. The former is quantified using the Consequence Reverse Diffusion Method (CRDM), which systematically identifies and categorizes key factors of primary dike failure modes into four domains: hydrological characteristics, channel morphology, engineering structures, and operational management. The latter is assessed by integrating socioeconomic impacts, including population exposure, infrastructure investment, and industrial and agricultural production. A structured assessment framework is established through systematic indicator selection, justified weight assignment, and standardized scoring criteria. Application of the framework to Yangtze-to-Huaihe Water Diversion Project (Henan Reach) reveals that the risk probability across four segments falls within the (1, 3) range, indicating a generally low to moderate risk profile, while channel morphology shows greater spatial variability than hydrological, structural, and management indicators, driven by local differences in crossing structure density, sinuosity, and regime coefficients. Meanwhile, the segments along the Qingshui River face higher risk losses owing to their upstream location and large-scale water supply capacity, resulting in a relatively higher comprehensive risk level.

In recent years, as the ocean absorbs increasing amounts of atmospheric CO2, ocean acidification has intensified. Simultaneously, global warming and enhanced ocean stratification have led to the continuous expansion of the oceanic oxygen minimum zone (OMZ). Under the combined effects of these processes, a key question arises: How will the transport of particulate organic carbon (POC) to the deep ocean (>1,000 m) be affected? Analysis of POC flux data from 547 stations, collected via global sediment traps since the 1980s, reveals that POC flux has increased only in the shallow ocean (<300 m) but has significantly decreased in the deep ocean. These findings suggest that the expansion of the OMZ has not led to more carbon being transported to the deep ocean. Instead, more POC is being retained in the mid-upper ocean (<1,000 m), where its degradation results in significant consumption of dissolved oxygen, contributing to the expansion of the OMZ.

The Indus River Basin is one of the most densely populated transboundary river basins in the world and is the region with the most serious water disputes. Given the current population's rapid growth, inclusive and high-resolution datasets are urgently needed to assess how this growth will affect the distribution of resources and the sustainability of the environment. Here we present a population gridded dataset for the Indus River Basin with a resolution of 2.5 arc-minutes (~5 km). Based on the historical population distribution and the provincial (state) demographic parameters of the four countries, Afghanistan, China, India, and Pakistan, we projected the population size and structure (age, sex) changes under the Shared Socioeconomic Pathways (SSP1-5) for the period of 2020-2100 on each grid in the Indus River Basin. The dataset was well verified by comparing it with the observed population gridded dataset in 62,140 grids in the basin. The dataset can be useful sources for further research in resource management, sustainable development initiatives, and assessment of climate change impact.

In the context of global climate change and urban expansion, urban residents are encountering greater rainstorm waterlogging risk. Quantifying population exposure to rainstorms is an important component of rainstorm waterlogging risk assessments. This study utilized a two-dimensional hydrodynamic model to simulate the inundation water depth and inundation area resulting from rainstorms, with return periods of 5, 10, 50, and 100 years, in the Xiong’an New Area, and overlaid the gridded population data in 2017 and in 2035 under SSP2 to assess the change in population exposure. The results show that the average inundation depth and area increase were from 0.11 m and 207.9 km2 to 0.18 m and 667.2 km2 as the rainstorm return period increased from once in 5 years to once in 100 years. The greatest water depths in the main urban areas were mainly located in the low-lying areas along the Daqing River. The total population exposed to rainstorm waterlogging for the 5-, 10-, 50-, and 100-year return periods was 0.31, 0.37, 0.50, and 0.53 million, respectively, in 2017. However, this is projected to rise significantly by 2035 under SSP2, increasing 2–4-fold compared with that in 2017 for the four return periods. Specifically, the projected population exposure is expected to be 0.7, 1.0, 1.8, and 2.0 million, respectively. The longer the return period, the greater the increase in population exposure. The proportion of the population exposed at the 0.05–0.2 m water depth to the total population exposure decreases as the return periods increases, whereas the proportion changes in the opposite direction at the 0.2–0.6 m and >0.6 m depth intervals. Spatially, high-exposure areas are concentrated in densely populated main urban regions in the Xiong’an New Area. In the future, more attention should be paid to densely populated low-lying areas and extreme recurrence rainstorm events for urban flood-risk management to ensure population safety and sustainable urban development.

High-quality ocean observation is essential for research and applications in ocean exploration and climate change. With moving into the era of big data in recent years, it becomes crucial to process these massive raw observations accurately and efficiently. This paper addressed issues encountered in processing ocean big data within traditional delayed-mode quality control systems, including substantial serial I/O workloads and frequent context switching. A parallel quality control scheme named CODC-pyParaQC was proposed by constructing computing process groups. It retains the advantages of the existed delayed-mode quality control system (e.g. CODC-QC) while improving the efficiency of the quality control procedure, solving the feasibility of a large-scale parallel computation of the quality control scheme and realizing the (near) real-time quality control of massive ocean observation profiles. The results showed that the efficiency of single-node quality control has been improved by about 10 times. Leveraging the computing power of supercomputers and employing multi process groups for cross-node parallel computation, we have developed a fast and efficient (near) real-time quality control procedure. This system processed approximately 22,548,733 temperature profiles from the world ocean database (1940-2023) in about 6.5 hours. Our new quality control scheme can ensure the computing capability necessary for establishing a high-quality ocean observation profile database.

Despite contributing less than 1% of global greenhouse gas (GHG) emissions, Small Island Developing States (SIDS) have the potential to drive global mitigation actions by advocating for ambitious emission reduction targets, promoting renewable energy solutions, and advancing sustainable development practices. The adoption of onshore-offshore wind and solar energy in 39 SIDS, which are currently experiencing the adverse effects of climate change, presents a significant opportunity. By harnessing renewable energy sources, these countries can effectively mitigate GHG emissions, enhance energy security, and build resilience. This approach aligns with the renewable energy roadmap outlined at the 28th Conference of Parties (COP) of the United Nations Framework Convention on Climate Change (UNFCCC), facilitating a transition from fossil fuels to renewable energy sources. However, realizing such prospects requires collaboration among policymakers, industry stakeholders, and researchers to address multiple technical, economic, and environmental issues. Through this joint effort, the untapped potential of wind and solar energy can be fully harnessed, offering a pragmatic solution to actively mitigate climate change and the issues faced in these regions.

In current research on the Anthropocene, assessing the impact of human activities via stratigraphic records of sediments and demarcating the Anthropocene epoch globally are critical scientific issues that urgently need to be addressed. The northeastern Qinghai-Xizang Plateau (QXP), where humans first settled permanently in large numbers in the QXP, has varying sedimentary environments that are extremely sensitive to human activities. In contrast to other regions of the QXP, the northeastern sector boasts a richer array of climatic and environmental reconstruction sequences. This distinctive feature renders it an exemplary locale for investigating the stratigraphic boundary of the Anthropocene. Through in-depth analysis and integration of existing paleoclimate and paleoenvironment sequences in the northeastern QXP, we draw the following conclusions: (1) Throughout the past millennium, the impact of human activities on the environment of the northeastern QXP has become increasingly significant, especially in the past 200–300 years, gradually overshadowing climatic factors. (2) Since AD 1950, multiple physicochemical indicators related to human activities in the northeastern QXP have shown exponential growth, forming a distinct peak within the past millennium and clearly depicting the global “Great Acceleration” phenomenon and its development process. (3) Intensified human activities have driven swift environmental shifts and “decoupled” the interplay between climatic variations and the ecological environment, propelling the northeastern QXP into the “Early Anthropocene” from the “Late Holocene”. On the basis of the above findings, we construct a model suitable for identifying the stratigraphic boundary of the Anthropocene in the northeastern QXP and note that since the ecological environment in the northeastern QXP has entered the “Early Anthropocene”, the climate signals of certain physicochemical indicators in sediments are gradually becoming weaker, whereas the signals of human activities are becoming stronger.

Climate change has remarkable global impacts on hydrological systems, prompting the need to attribute past changes for better future risk estimation and adaptation planning. This study evaluates the differences in simulated discharge from hydrological models when driven by a set of factual and counterfactual climate data, obtained using the Inter-Sectoral Impact Model Intercomparison Project's recommended data and detrending method, for quantification of climate change impact attribution. The results reveal that climate change has substantially amplified streamflow trends in the Upper Yangtze and Upper Yellow basins from 1961 to 2019, aligning with precipitation patterns. Notably, decreasing trends of river flows under counterfactual climate have been reversed, resulting in significant increases. Climate change contributes to 13%, 15% and 8% increases of long-term mean annual discharge, Q10, and Q90 in the Upper Yangtze at Pingshan, and 11%, 10%, 10% in the Upper Yellow at Tangnaihai. The impact are more pronounced at headwater stations, particularly in the Upper Yangtze, where they are twice as high as at the Pingshan outlet. Climate change has a greater impact on Q10 than on Q90 in the Upper Yangtze, while the difference is smaller in the Upper Yellow. The impact of climate change on these flows has accelerated in the recent 30 years compared to the previous 29 years. The attribution of detected differences to climate change is more obvious for the Upper Yangtze than for the Upper Yellow.