Search Results (1,309)

Anomaly detection is critical in safety-sensitive fields, but faces challenges from scarce abnormal data and costly expert labeling. Time series anomaly detection is relatively challenging due to its reliance on sequential data, which imposes high computational and memory costs. In particular, it is often composed of real-time collected data that tends to be noisy, making preprocessing an essential step. In contrast, image anomaly detection has leveraged advancements in technologies for analyzing spatial patterns and visual features, achieving high accuracy and promoting research aimed at improving efficiency. We propose a novel framework that bridges image anomaly detection with time series data. Using Gramian Angular Field (GAF) transformations, we convert time series into images and apply state-of-the-art techniques, Reverse Distillation (RD) and EfficientAD (EAD), for efficient and accurate anomaly detection. Tailored preprocessing and transformations further enhance performance and interoperability. When evaluated on the multivariate time series anomaly detection dataset Secure Water Treatment (SWaT) and the univariate datasets University of California, Riverside (UCR) and Numenta Anomaly Benchmark (NAB), our approach demonstrated high recall overall and achieved approximately 99% F1 scores on some univariate datasets, proving its effectiveness as a novel solution for time series anomaly detection. Full article

Geographic data are the foundation of geographic model construction, and any stage of their acquisition, processing, and analysis may have an impact on the efficiency and quality of geographic modeling and simulation. With the advent of the era of big data, a large number of data resources are generated in the field of geographic information. However, due to the heterogeneity of geographic data and the security of data usage, massive geographic data resources are difficult to fully explore and utilize, resulting in the formation of data islands. This paper proposes a service-oriented geographic data-sharing and computing framework, which provides users with a complete set of geographic data access and application processes (such as data acquisition, processing, configuration, etc.), so as to reduce the difficulty of using data and improve the efficiency of data sharing. The framework mainly consists of three core components: (1) the “Data service container” can publish data resources as data services to provide a consistent data access interface; (2) the “Workspace” provides a series of methods and tools for users to develop data-computing solutions; and (3) the “Data-computing engine” is responsible for performing computing tasks such as data processing and configuration. Finally, a case of runoff simulation using the SWAT model is designed, in which the whole process of data sharing, acquisition, calculation, and application is realized, so as to verify the validity of the proposed framework. Full article

Analysing the patterns and impacts of land-use changes in the production–living–ecological space (PLES) of the Fenhe River Basin (FRB 39,721 km²), China, is necessary to support sustainable development. Based on remote sensing images from 1990 to 2020, we aimed to analyse the PLES land-use changes. Industrial production and living spaces continuously encroached on the agricultural production and ecological spaces between 1990 and 2022 owing to industrialisation and urbanisation, and the ecological land area decreased by 699.21 km², while the industrial production land area increased by 521.32 km². We used the soil and water assessment tool (SWAT) model to quantitatively analyse the impact of PLES changes on runoff in the FRB. With the continuous expansion of production and living spaces, the extensive use of concrete in cities has led to ground hardening, making it difficult for precipitation to infiltrate, with surface runoff increasing by 0.3 mm annually. The reduction in ecological space has led to a reduction in forests and grasslands, weakening the water-holding capacity of the watershed and affecting groundwater storage. This study provides a scientific basis for watershed management and the integrated development of PLES. Full article

This study aimed to evaluate the impact of a wildfire on vegetation recovery and hydrological processes in a Mediterranean peri-urban system, using remote sensing and hydrological modeling. NDVI and MSAVI₂ time series extracted from burned areas, control plots, and VAR-modeled plots were used to analyze vegetation regeneration. The SWAT model, calibrated for pre-fire conditions due to data limitations, was used to evaluate subbasin-scale hydrological impacts. Results showed limited recovery in the first post-fire year, with vegetation indices remaining lower in burned areas compared to control plots. High- and moderate-burn-severity areas presented the most significant NDVI and MSAVI₂ increases. The SWAT model showed increased water yield, percolation, and surface runoff with reduced evapotranspiration in post-fire conditions. Peak discharges were notably higher during wet periods. Modified land use and soil properties affected the catchment’s hydrological balance, emphasizing the complexities of post-fire catchment dynamics. Full article

Synthetical eco-environmental problems’ treatment is a new stage for certain pollutant control or ecological restoration. Traditional urban planners have focused more on social–economic development but less on eco-environmental considerations. Spatial planning is currently an essential administrative management method for regional development and eco-environmental protection in China. National and provincial spatial planning designs general strategies, and prefecture-level planning is the most important scale for spatial management. For scientific, spatial governance for eco-environmental protection, we propose a synthetic spatial analysis and planning method framework that involves atmospheric, edaphic, hydrographic, and ecological processes to identify pivotal regions for regional eco-environmental security goals. The synthetic method was conducted using advanced models including the CMAQ and SWAT models and spatial statistical methods. A Chinese prefecture-level city, Anshan City, was chosen to fulfill the method framework due to its various ecosystem types and environmental problems. A total of 67 eco-environmental management units (EMU) were divided based on atmospheric pollution patterns, hydrographic processes, edaphic heavy metal pollution, and ecological spatial analysis. Each unit was identified with ecological or environmental risk and a proposed management regulation. For considering the whole eco-environmental process, the ecological security pattern (ESP) was constructed. The results showed that 166 corridors were identified with an area of 2241.25 km², with enhanced connectivity among 76 ecological sources (12.27% of Anshan City). By coupling two results, the optimized ecological conservation and restoration pattern was proposed, in which priority protection areas were identified. This synthetic method can provide scientific analysis and guidance to support spatial planning and ecological construction for multi-purpose ecological and environmental protection. Full article

Climate change and human interventions have exerted a long-term influence on variations in continental streamflow. Despite this, the precise mechanisms by which these factors regulate the change in streamflow remain inadequately understood, especially in arid alpine regions, due to the limited number of observations which exacerbates difficulties in comprehensively assessing streamflow alterations. Consequently, assessing the impacts of climate change and human interventions on streamflow is a challenge in data-scarce regions. Here, using the Zamu River as an example, we analyzed streamflow changes in arid alpine regions using a method that integrates the Patch-generated Land Use Simulation model, the Soil and Water Assessment Tool, and the Coupled Model Intercomparison Project Phase 6. Our analysis highlighted that climate change primarily drove streamflow variations in the Zamu River, accounting for over 80% of the observed contributions. This influence was further amplified by the effects of future climate and changes in land use and land cover, resulting in increased streamflow. Additionally, precipitation emerged as the main factor driving the rise in streamflow. These findings emphasize the significant impact of climate change on water cycles in arid alpine regions and underscore the necessity for tailored water resource management strategies to ensure sustainable regional development and effective climate change adaptation. Full article

Multi-purpose dams in a river basin frequently result in variations in downstream flow. Precisely assessing the reservoir operation effects can improve management strategies and alleviate extreme hydrological events. This study assesses the impact of reservoir operation scenarios on the downstream flow in the Seomjin River basin in South Korea. Four reservoir scenarios were developed utilizing observed daily inflow and outflow data from the reservoirs. A semi-disturbed hydrological model, SWAT (Soil and Water Assessment Tool), was employed to simulate the flow for each reservoir operation scenario in the downstream section of the study basin. Model execution was evaluated by comparing the simulated and measured streamflows using performance metrics, including R², NSE, and PIBAS, which displayed very good compatibility. The sensitivity of calibration parameters varied across different reservoir operation scenarios. The results of this study indicate that the operation scenarios for the Seomjin and Juam reservoirs led to a maximum downstream flow reduction of 32%. Additionally, the monsoon season exhibited a lower percentage reduction in flow compared to the dry season, which was influenced by the frequency of rainfall in the region. Annual assessment indicated that streamflow reduction varies between 1.35% and 32.9% across all reservoir operation scenarios. Reservoir operations have demonstrated their effect on the alteration of downstream flow in the Seomjin River basin. This study demonstrates that the operation of the Seomjin reservoir has a more significant impact on downstream flow than that of the Juam reservoir in the study region. This study analyzed a substantial basin with various reservoir operation scenarios to assess the influence of flow on the downstream section, yielding important insights for efficient water resource management. Full article

The Yarlung Zangbo River (YZR) is a sizeable highland river on the Tibetan Plateau, and its runoff process is crucial for understanding regional water resource features and related ecological patterns. However, the runoff characteristics of the YZR Basin (YZRB) remain unclear, especially how it would react to climate change. This study comprehensively analyzed the runoff characteristics of the entire YZRB based on a validated distributed hydrological model (SWAT) coupled with a glacier module (SWAT-glac), identified the runoff components, and explored the climate–discharge relationship, with a particular focus on the relationships between glacier runoff and changes in precipitation and air temperature. The results indicate that the SWAT-glac model, with localized glacier parameters, accurately simulates the runoff processes due to regional differences in meteorological conditions and uneven glacier distribution. Summer runoff dominates the basin, contributing 46.2% to 57.9% of the total, while spring runoff is notably higher in the downstream sections than in other areas. Runoff components vary significantly across river sections; precipitation is the primary contributor to basin-wide runoff (23.4–59.5%), while glacier runoff contribution can reach up to 54.8% in downstream areas. The study found that underlying surface conditions, particularly glacier coverage, significantly influence runoff responses to meteorological changes. The correlation between runoff and precipitation is stronger at stations where rainfall predominates, whereas runoff shows greater sensitivity to air temperature in glacier-covered areas. These findings enhance the understanding of runoff processes in the YZRB and offer valuable insights for the sustainable management of water resources in similar basins under climate change. Full article

Adversarial attacks targeting industrial control systems, such as the Maroochy wastewater system attack and the Stuxnet worm attack, have caused significant damage to related facilities. To enhance the security of industrial control systems, recent research has focused on not only improving the accuracy of intrusion detection systems but also developing techniques to generate adversarial attacks for evaluating the performance of these intrusion detection systems. In this paper, we propose a deep reinforcement learning-based adversarial attack framework designed to perform man-in-the-middle attacks on industrial control systems. Unlike existing adversarial attack methods, our proposed adversarial attack scheme learns to evade detection by the intrusion detection system based on both the impact on the target and the detection results from previous attacks. For performance evaluation, we utilized a dataset collected from the secure water treatment (SWaT) testbed. The simulation results demonstrated that our adversarial attack scheme successfully executed man-in-the-middle attacks while evading detection by the rule-based intrusion detection system, which was defined based on the analysis of the SWaT dataset. Full article

The CE-QUAL-W2 model is a significant tool extensively used in lentic environments to analyze eutrophication and water quality. This systematic review of the CE-QUAL-W2 hydrodynamic model revealed its widespread application in analyzing reservoir eutrophication. A total of 151 relevant papers were identified, of which 38 were selected after rigorous analysis, showcasing studies in environmental sciences and water resources. In 2021, we saw the highest number of publications, with six papers; 2022 achieved the highest number of citations, with 113. The model has been widely used across countries, with Iran leading in the number of publications, followed by China and Brazil. The standard combination of CE-QUAL-W2 with the SWAT model reflects its effectiveness in complex watershed studies. CE-QUAL-W2 has demonstrated the ability to predict future environmental conditions and diagnose environmental extremes, and it can calculate various hydrodynamic and water quality parameters. Its increasing use in high-impact scientific journals underscores its global relevance and particular promise for Brazilian aquatic environment studies due to its efficiency and accessibility. With its significant potential, this model is poised to enhance the understanding and management of water resources, contributing to environmental sustainability and inspiring optimism for future applications on a global scale. Full article

The gullies and valleys of the Loess Plateau, as key ecological zones for soil erosion control, play a critical role in the region’s sustainable development under increasing urbanization. This study employed the Soil and Water Assessment Tool (SWAT) to analyze the impacts of land use/cover changes (LUCC) on runoff at multiple spatial scales and locations within the Sanchuan River Basin (SRB) in the loess hilly and gully region. The methodology integrates SWAT modeling with LUCC scenario analysis, focusing on spatial and scale effects of land use changes on hydrological processes. The results revealed distinct spatial differences, with diminishing LUCC impacts on streamflow from the upper to lower reaches of the basin, regardless of land use type. Scale effects were also evident: grassland effectively controlled runoff within 300 m of riparian zones, while forest land was most effective beyond 750 m. A relatively insensitive range for runoff changes was observed between 300 and 750 m. These findings highlight the critical role of LUCC in influencing runoff patterns and underscore the importance of region-specific and scale-sensitive land use management strategies. This research provides valuable guidance for sustainable land planning, particularly in riparian zones, to enhance runoff control and optimize ecological benefits. Full article

Runoff research serves as the foundation for watershed management, and the relationship between runoff and landscape pattern represents a crucial basis for decision-making in the context of watershed ecological protection and restoration. However, there is a paucity of research investigating the multi-scale spatial relationship between runoff and landscape patterns. This study employs the Poyang Lake Basin (PLB) as a case study for illustrative purposes. The construction of the soil and water assessment tool (SWAT) model is the initial step in the process of carrying out runoff simulation, which in turn allows for the analysis of the spatial–temporal characteristics of runoff. Subsequently, Pearson’s correlation analysis, global linear regression and geographically weighted regression (GWR) models are employed to examine the impact of landscape composition on runoff. Finally, the spatial relationship between runoff and landscape pattern is investigated at the landscape and class scales. The results of the study demonstrate the following: (1) runoff in the PLB exhibited considerable spatial–temporal heterogeneity from 2011 to 2020. (2) Forest was the most prevalent landscape type within the PLB. Landscape composition’s impact on runoff exhibited non-linear characteristics, with forest, cropland, barren, and grassland influencing runoff in decreasing order. (3) A spatial relationship between runoff and landscape pattern was observed. At the landscape scale, patch diversity significantly influenced runoff, and reducing patch diversity primarily increased runoff. At the class scale, forest and cropland patch areas had the greatest impact on runoff, potentially enhanced by improving patch edge density. (4) Nine sub-basins needing ecological restoration were identified, with restoration pathways developed based on spatial relationships between runoff and landscape patterns. This study elucidates the impact of landscape composition and pattern on runoff, thereby providing a basis for informed decision-making and technical support for the ecological restoration and management of the watershed. Full article

Understanding hydrological processes is crucial for effective watershed management, with SWAT+ being one of the widely adopted models for analyzing water balance at watershed scales. While hydrological components are often assessed through sensitivity analysis, calibration, and validation, parameter sensitivity during dry periods (low-flow conditions) when baseflow is predominant remains a relevant focus, especially for watersheds like Majalaya, Indonesia, which experience distinct low-flow periods. This study analyzes water balance components in the Majalaya watershed, Indonesia, using SWAT+ across the 2014–2022 period, focusing on low-flow conditions. This study employs a two-step calibration approach using various datasets, including ground rainfall (2014–2022), NASA POWER meteorological data, MODIS land cover, DEMNAS terrain, and DSMW soil types, and the streamflow data for model calibration. The first calibration step optimized the overall performance (R² = 0.41, NSE = 0.41, and PBIAS = −7.33), and the second step improved the baseflow simulation (R² = 0.40, NSE = 0.35, and PBIAS = 12.45). A Sobol sensitivity analysis identified six primary parameters, i.e., CN3_SWF, CN2, LATQ_CO, PERCO, SURLAG, and CANMX, as the most influential for streamflow calibration, with CN3_SWF and CN2 being the most critical. This study demonstrates SWAT+’s effectiveness in watershed management and water resource optimization, particularly during low-flow conditions. Full article

The water level and surface area of Urmia Lake, located in the northwest of Iran, has decreased dramatically, presenting significant challenges for hydrological modeling due to complex interactions between surface and groundwater. In this study, the impact of agricultural activities on streamflow within one of the largest sub-basins of Urmia Lake is assessed using the Soil and Water Assessment Tool (SWAT) for hydrological assessments. To have accurate assessments, land use change detections were considered by a novel method, which merges the Normalized Difference Vegetation Index (NDVI) with the Digital Elevation Model (DEM) to create a two-band NDVI-DEM image, effectively differentiating between agricultural and rangeland fields. Our findings reveal that agricultural development and irrigation, escalating between 1977 and 2015, resulted in increased annual evapotranspiration (ET) (ranging from 295 mm to 308 mm) and a decrease in yearly streamflow, from 317 million cubic meters to 300 million cubic meters. Overall, our study highlights the significant role that agricultural development and irrigation may play in contributing to the shrinking of Lake Urmia, underscoring the need for improved regional water management strategies to address these challenges, though further analysis across additional basins would be necessary for broader conclusions. Full article

With the intensification of climate change and human activities, the impacts of land use shifts on hydrological processes are becoming more pronounced, especially in regions with complex geographic, geological, and climatic conditions such as the Northeast Black Soil Region, China. This study quantitatively examines the variations in various land use types from 1980 to 2020 by means of a land use transfer matrix, and it incorporates the multi-year average runoff value to mitigate the interference of short-term climate fluctuations on the runoff trend, thereby enhancing the representativeness and stability of the simulation outcomes. The SWAT (Soil and Water Assessment Tool) model is employed to simulate land use alterations in different periods. The findings indicate that the area of farmland increased by 5.34% and the area of grassland decreased by 5.36% over 40 years. The areas of forest land and wetland have fluctuated significantly due to policy interventions and population growth. This study discovers that LUCC has resulted in a marginal increase in annual water yield. For instance, the water yield of paddy fields in 2020 amounts to 92.26 mm/year, which is 0.52–9.42% higher than the historical scenario and exhibits a notable upward trend in summer. Spatial analysis discloses regional disparities, with substantial changes in the hydrological behavior of northern watersheds (such as the Huma River) and southeastern regions (such as the Toudao River). The augmentation of wetland and forest coverage has effectively mitigated peak runoff, especially during extreme rainfall events. Wetlands have manifested strong water regulation capabilities and alleviated the impact of floods. This study quantitatively discloses the complex response pattern of LUCC to runoff by introducing a multi-scale analysis approach, which furnishes a scientific basis for flood risk assessment, land use optimization, and water resource management, and demonstrates the potential for extensive application in other countries and regions with similar climatic and topographic conditions. Full article