Water Resources Carrying Capacity (WRCC) embodies the fundamental constraint relationship between water resource sustainability and regional economic growth as well as social development, serving as a key indicator for assessing regional sustainable development. Existing studies mainly focus on carrying capacity or status assessment, often deriving deterministic results based solely on a single factor. However, as an indicator of sustainability, WRCC is inevitably influenced by uncertainty risks. Therefore, This study develops a novel Water Resources Carrying Capacity Early Warning (WRCCEW) system based on the Driving force-Pressure-State-Impact-Response-Management (DPSIRM) framework and applies it to evaluate the status of Henan Province from 2011 to 2023.Using the Geographical Detector (GD) model, this study examined the drivers of WRCCEW during different periods along with their interaction effects, while the Bayesian Network (BN) model was applied to simulate the probability of risk occurrence under diverse scenarios. Findings reveal that in Henan Province, the no warning zones are migrating from the west toward the south, the proportion of extreme warning zones has markedly decreased, whereas the proportion of no warning zones has experienced a modest reduction. Meanwhile, the interactive effects of multiple factors exert a significantly stronger influence on WRCCEW than single-factor effects, with the coupling between total water resources and other factors being the most prominent. In the scenario simulations, the S18 scenario can significantly reduce the occurrence of high-risk situations in Henan Province. This study provides a comprehensive framework for the early risk assessment of regional WRCC, highlighting the importance of incorporating uncertainty and probabilistic risks in sustainability evaluation.

Achieving model interpretability and high-precision multi-step-ahead daily streamflow prediction remains a challenge in data-limited watersheds. First, causal learning is performed to quantify the causal strength of different factors, based on the causal-oriented representation learning predictor (CReP). Second, the multi-head self-attention mechanism integrated with causal information (CI-MHSA) is developed to enable efficient fusion of spatiotemporal features, and then a coupled framework called CI-MHSA-CReP is proposed. CI-MHSA-CReP is applied to 5–15-day-ahead streamflow prediction at four stations with a training dataset length of 1200. It is found that the predicted values generated by CI-MHSA-CReP exhibit excellent agreement with the observed values during low-flow, normal-flow, and high-flow periods, with minimal time-lag effects. Meanwhile, the model delivers highly consistent prediction performance across 5-step, 10-step, and 15-step forecasting. The average NSE values of CI-MHSA-CReP reach 0.830, 0.713, and 0.647 for 5-step, 10-step, and 15-step predictions respectively, representing an improvement of 4.3–75.8

Accurate and reliable multi-step ahead streamflow forecasting is important for water resource management and flood prevention. To alleviate the temporal lag in multi-step prediction and improve peak prediction capability, this research develops a multi-head self-attention-spatiotemporal skip-connection model (MHSA-STSM), which is based on nonlinear dynamic systems and deep learning approaches. MHSA-STSM comprises a temporal module constructed from a convolutional neural network (CNN), a spatiotemporal module fashioned from a multi-head self-attention mechanism, along with a skip connection that links directly to the original input; these modules enable MHSA-STSM to effectively amalgamate temporal, spatiotemporal, and global information within the data. By learning the mapping between the original attractors and the delay attractors, MHSA-STSM can extract spatiotemporal features from the original attractors, thereby enabling the prediction of future values for the target variable. MHSA-STSM is applied to make a multi-step forecast of daily streamflow in rivers in the states of Maine, USA. For a five-step forecast, the highest R value of MHSA-STSM is 0.960, which is 1.05%-11.10% higher than CNN, multi-head self-attention mechanism-Long Short-Term Memory (MHSA-LSTM) and STSM; the lowest R value of 0.792 is at the USGS1047000 station, which shows a 91.4% improvement over the average of CNN, MHSA-LSTM and STSM; the RMSE and MAPE values of MHSA-STSM are 10.76%-102.50% and 19.26%-305.51% lower than those of three comparative models; the NSE of MHSA-STSM is significantly greater than that of the other models, and is as high as 0.920 at USGS 01013500 station. Moreover, sensitivity experiments on the prediction step length are performed for the model. It is found that MHSA-STSM performed excellently in five-step, seven-step, and ten-step predictions and can effectively alleviate the time lag issue. The R value ranges from 0.960 to 0.938, with NSE from 0.920 to 0.836. As the step length increases from 5 to 10, the R value decreases by only 2.3%, and the NSE decreases by 9.1%, demonstrating high stability, while the performance of other models significantly declines. Therefore, MHSA-STSM can effectively capture the spatiotemporal information embedded in high-dimensional data and make accurate multi-step predictions of daily streamflow.

The sustainable utilization of water resources plays a crucial strategic role in regional economic development. The water resources carrying capacity (WRCC) is a multifaceted system influenced by diverse factors, where the interplay among water resources, societal factors, economic conditions, and ecological elements collectively determines the overall WRCC. Combining relevant research results, this paper utilized an improved TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) and GRA (grey relational analysis)-based WRCC evaluation model, introduced the panel vector autoregressive (PVAR) model to analyze the effects of interactions among subsystems, and applied the geographically and temporally weighted regression (GTWR) model for the driving analysis of WRCC. Using Ningxia Hui Autonomous Region as a case study, this paper discusses the internal dynamic relationships and driving mechanisms of the WRCC system. It also provides a new perspective for discussing WRCC in water-scarce areas and provides novel approaches for optimizing water resource management and enhancing ecological protection. The results indicate that the water resources subsystem is central to the WRCC in Ningxia, with significant interconnections among the four subsystems. However, significant spatial and temporal heterogeneity is evident across different regions. The water resources system contributes significantly, with ecological development having a positive impact on water resources. However, social and economic development has a restrictive impact on water resources.

Making accurate and reliable predictions for monthly runoff in data scarce basins is still a major challenge. In this study, a new model, the CL-NDM, is developed by combining Convolutional Neural Network-Long Short-term Memory (CNN-LSTM) and a nonlinear dynamic model. The CL-NDM can overcome the deficiency of observed data by fusing spatial and temporal dependencies in runoff sequences at different stations. First, phase space reconstruction is used to enlarge the dimensions of the runoff sequences and reconstruct the attractors of the runoff sequences. Then, the CNN-LSTM is employed to construct the mapping between non-delay and delay attractors. Finally, the prediction set of the target variable is obtained by embedding multiple times. The CL-NDM is performed for monthly runoff prediction at eleven hydrological stations in the Weihe River, China. Compared with the CNN, LSTM and CNN-LSTM models, which require a large amount of training samples, the CL-NDM behaves much better, especially in situations with small training sample sizes. The maximum increase in R is 74

It is crucial yet challenging to estimate the parameters of hydrological distribution for hydrological frequency analysis when small samples are available. This paper proposes an improved Bootstrap and combines it with three commonly used parameter estimation methods, i.e., improved Bootstrap with method of moments (IBMOM), maximum likelihood estimation (IBMLE) and maximum entropy principle (IBMEP). A series of numerical experiments with different small sized (10, 20, and 30) of samples generated from the three commonly used probability distributions, i.e., Pearson Type III, Weibull, and Beta distributions, are conducted to evaluate the performance of the proposed three methods compared with the cases of conventional Bootstrap and without-Bootstrap. The proposed methods are then applied to the estimation of distribution parameters for the average annual precipitations of 8 counties in Qingyang City, China with assumption of Pearson Type III distribution for the average annual precipitations. The resulting absolute deviation (AD) box plots and Root Mean Square Error (RMSE) and bias estimators from both the numerical experiments and the case study show that the estimated parameters obtained by the improved Bootstrap methods have less deviation and are more accurate than those obtained through conventional Bootstrap and without-Bootstrap for the three distributions. It is also interestingly found that the improved Bootstrap provides more relative improvement on the parameter estimation when smaller size of sample is used. The method based on improved Bootstrap paves a new way forward to alleviating the need of large size of sample for quality hydrological frequency analysis.

早在 20 世纪 70 年代黄河上游实测输沙量就开始大幅减少,其中黄河下河沿站 1970-1999 年和 2000-2022 年年均输沙量分别较天然时期减少 50%、76%,但迄今有关该区沙量锐减的研究极少.以黄河李家峡大坝至青铜峡区间的黄土高原为重点,分析了该区水沙和有效降雨变化情况、水库和淤地坝在不同时期的拦沙作用,还原了 1960 年以来流域实际产沙量,研究了梯田和植被变化情况,计算了现状下垫面在设计降雨情景下的产沙量.研究结果表明:20 世纪后期水库拦沙是研究区输沙量大幅减少的主要原因,2000 年以来梯田建设和植被改善逐渐成为主要减沙因素、水库拦沙次之;基于 2021 年下垫面推算的下河沿以上产沙能力为5 800万～6 600 万t/a、较天然时期降低约 62%,清水河流域产沙能力为1 920 万～2 743 万t/a、较天然时期降低约53%;未来植被改善潜力很小,新建梯田仅可弥补老旧梯田田埂损毁的负面影响,因此产沙能力进一步降低的潜力极小.

为助力黄河流域国家中心城市水资源高效利用和经济社会高质量发展,基于超效率SBM模型、Malmquist指数和Tobit回归模型,对郑州市和西安市水资源利用、社会经济和生态环境 3 个子系统的水资源利用效率、年际变化和影响因素进行分析,结果表明:1)2005-2020 年郑州市和西安市水资源利用效率适中,水资源利用子系统和社会经济子系统,郑州市的水资源利用效率高于西安市;生态环境子系统,西安市的水资源利用效率高于郑州市;郑州市的综合水资源利用效率高于西安市.2)郑州市 3 个子系统的全要素生产率(Malmquist指数)整体处于上升状态,西安市社会经济子系统的全要素生产率整体处于上升状态、水资源利用子系统和生态环境子系统则处于下降状态.3)经济发展水平与水资源利用效率成较为显著的正相关关系,产业结构、科技创新水平与水资源利用效率相关性不显著,水资源禀赋与水资源利用效率负相关.

水文序列的变异诊断研究能够为各类水利、土木工程规划和管理决策提供依据和参考.针对水文序列动力学结构变异难以诊断的问题,本文提出一种基于矩阵Renyi α阶熵的变异诊断方法.首先,引入矩阵Renyi α阶熵理论描述水文序列的动力学结构;其次,利用数据滑动技术构造滑动移除矩阵Renyi α阶熵序列,用来刻画水文系统动力学结构的演变;最后,利用Pettitt检验诊断出滑动移除矩阵Renyi α阶熵序列的变异点及显著性水平.以渭河流域、洮河流域、窟野河流域和西柳沟流域的水沙序列为例开展应用研究,并与Shannon熵、Mann-Kendall检验和滑动T检验等方法进行对比分析.研究结果表明:咸阳站、状头站和红李区间年径流序列均没有发生变异,其它站年径流序列均发生了变异,变异概率大于90％;咸阳站、张家山站和状头站年输沙量序列均在20世纪80年代发生了变异,其它站点在20世纪90年代末发生变异,变异发生的概率均大于95％.通过与现有研究成果比对分析发现,本文提出的Renyi α阶熵方法的诊断结果与实际情况基本吻合,而其它方法的诊断结果与实际情形差异较大.

为更好地了解采煤扰动下潜水位及包气带水分变化规律,在陕北典型矿区开展了降雨、潜水位、包气带土壤含水率等水循环要素的野外原位观测试验,基于观测数据,采用Spearman秩相关系数检验、小波分析等方法,分析了未开采区及采空区潜水位和包气带水分的变化特征.结果表明:未开采区地下水位对于降水的响应明显且时间上存在4、5个月的滞后,采煤扰动后,地下潜水位持续下降,与降水响应关系微弱;在垂向上,未开采区较大降水可对100 cm以下埋深的土壤含水率产生影响,采空区土壤含水率总体减小,且同降水的响应程度不显著,含水率最大值相对于未开采区出现时间提前,50 cm以下埋深的土壤含水率对小强度降水无响应.采煤扰动潜水位下降后造成包气带增厚,包气带损耗的水量增加,随之造成降雨入渗补给地下水减少,进一步加剧了潜水位下降.

归一化植被指数(NDVI)可以有效地反映地表植被的生长状况,研究植被NDVI空间分布的驱动因素有助于区域生态环境保护.该研究基于2000—2018年黄河源区MODIS-NDVI数据和同时期8种自然因子数据,运用趋势分析法分析黄河源区植被NDVI时空变化特征,并利用地理探测器分析其空间分布的自然驱动因子.研究表明,黄河源区植被覆盖总体较高,2018年区域内74％的面积NDVI大于0.6.NDVI分布特点为东南高西北低,变化格局为北部增加,中部减少.2000—2018年NDVI均值总体上呈增加趋势,但变化趋势不明显,增长率为0.013/10 a;除高植被覆盖区面积增加外,其他等级植被覆盖区面积均减小.年降水量对NDVI空间分布的影响力最大,达到0.602;高程影响力为0.385,年均温影响力为0.296,也很好地解释了黄河源区的植被覆盖状况;其他自然因子对NDVI空间分布的影响力较小.自然因子对植被NDVI的影响呈现相互增强和非线性增强关系,使地貌类型、坡度及坡向等影响较小的单因子对植被NDVI也有了较大的影响,其中,年降水量与其他因子之间的交互作用的影响力普遍较高,年降水量与海拔的交互作用影响力最大,达到0.682.研究表明,2000—2018年黄河源植被覆盖度呈不显著增加,年降水量是影响植被NDVI空间分布的主导因子,自然因子对植被NDVI的影响具有交互作用.该研究有助于更好地认识黄河源区植被覆盖情况以及植被空间分布的影响机制.

Encounter risk precipitation of rich-poor precipitation is beneficial for the utilization of flood resources and rational allocation of water resources which often involves a challenging task—estimating the joint probability distribution function (PDF) of multiple hydrologic variables using copulas. This paper introduced a linear combination of three copulas (combined copula) to study probabilistic characteristics of precipitation in two watersheds. To validate the performance of the combined copula, four experiments were employed to identify the joint distribution for the summer monthly precipitation and annual precipitation at two pairs of neighboring stations in Jinghe River, China, which were then compared with three individual copulas, namely, Gumbel copula, Clayton copula, and Frank copula. All the experiments showed that the combined copula performed much better than any of the three individual copulas. The combined copula was further applied to predict the synchronous-asynchronous probabilities of the summer monthly precipitation and annual precipitation at those four stations in Jinghe River. The rich-normal-poor synchronous encounter probabilities of the summer monthly precipitation reach up to 0.7 and 0.63 for Guyuan-Pingliang stations and Huanxian-Xifeng stations, respectively. The rich-normal-poor synchronous encounter probabilities of the annual precipitation reach up to 0.6 and 0.59 for the Guyuan-Pingliang stations and the Huanxian-Xifeng stations, respectively. Moreover, the encounter probability of rich-poor precipitation between receiving areas of Haihe River and upper reaches of Han River was calculated by the combined copula, and the probability that is suitable to transfer water is about 0.35.

植被作为反映陆地生态系统和气候的重要指标,对研究全球或区域生态环境变化具有重要作用.以地处黄土高原生态脆弱区的榆林市为研究区,基于地理探测器模型,选取坡向、坡度、气温、降水和土壤类型5类自然因子,土地利用类型、人口密度和GDP 3类人文因子,分析榆林地区植被空间分异特征及其驱动力,并揭示了促进植被生长影响因子的最适宜特征.结果表明,(1)研究区2000—2018年植被覆盖趋向改善,NDVI呈现增加趋势,增速为0.11/10 a,2008年以后植被增长较为明显;NDVI在2018年中高等级(0.6—0.8)面积比2000年中高等级面积明显增加;中高等级集中于榆林市东部黄土丘陵区,中低等级(0.2—0.4)集中于榆林市西北部的风沙区,植被覆盖呈现东部高西北低的空间分布特征.(2)人口密度和气温因子较好地解释植被NDVI空间分异性,是影响NDVI空间分异性的主要因子,GDP、土地利用类型和坡度是次级影响因子,其他因子对NDVI空间分异存在间接影响;坡向、降水和土壤类型因子与其他自然、人文因素对植被空间分布影响存在显著性差异.(3)自然、人文因子对榆林市NDVI的影响存在交互作用,因子之间的交互效应表现为相互增强或非线性增强关系,不存在独立关系.该研究揭示了促进植被生长的各影响因素适宜类型或范围,自然、人文因子的共同作用对植被影响更加显著,为地方政府指导区域植被恢复和生态修复提供科学依据.

为了揭示自然因子对三江源植被覆盖度的影响,基于1982-2015年三江源GIMMS NDVI数据和同时期8种自然因子,运用一元线性回归法分析三江源NDVI时空变化特征,并利用地理探测器分析了其空间分异性及自然驱动因素.结果表明:三江源植被NDVI分布东南高西北低,变化格局为西部和北部增加,中部和南部减少,1982-2015年NDVI均值总体上呈增加趋势;年降水量对三江源NDVI空间分布的影响力最大(0.551),年均温、植被类型、高程也很好地解释 了三江源的植被覆盖状况;自然因子对NDVI的影响存在交互作用,呈现相互增强和非线性增强关系.可见,年降水量是影响三江源NDVI空间分布的最主要因子,并且与其他因子交互作用能够增大对NDVI的影响.

: The Yellow River Basin is mainly distributed in arid, semi-arid, and semi-humid areas with a fragile ecological environment, which has become the most serious soil erosion area in China and even in the world. Evapotranspiration (ET) is an essential part of the land water and energy cycle, playing a vital role in the global ecosystem regulation and the hydrological cycle. ET is mainly composed of vegetation and transpiration of soil water and different vegetation cover or land-use types, showing different spatial-temporal distribution characteristics. Accurate exploration of the ET ’ s spatial-temporal distribution and its response relationship is conducive to mastering regional surface water and heat balance law. Remote sensing has been widely used in ET ’ s dynamic monitoring on a global or regional scale. Analyzing the ET ’ s spatial-temporal changes in the Yellow River Basin promotes understanding the impact of vegetation and land-use changes on the water cycle and the rational allocation of water resources. On the basis of ET- MODIS data, normalized difference vegetation index (NDVI), and land- use products as data sources, we studied the ET ’ s spatial-temporal change pattern in the Yellow River Basin from 2001 to 2015 using the Manner- Kendall test and Sen ’ s trend analysis. We discussed ET ’ s impact and the different change characteristics under different NDVI conditions and land- use types. We found the following results:

Prediction of water shortage losses is of great importance for water resources management. A new mathematical expression of water shortage loss was proposed in order to describe the random uncertainty and economic attributes of water resources. Then, Gumbel copula with a new method of parameter estimation was introduced to model the joint probabilistic characteristics for water supply and water use in situations when sufficient data is unavailable. The new parameter estimation method requires only the minimum and maximum values of two variables. The improved Gumbel copula was proved to be reliable based on the RMSEs (root mean square error) and AICs (Akaike information criterion), statistical tests and upper tail dependence tests. The potential water shortage losses for all the districts of Tianjin were predicated. The water shortage loss in the Urban district is highest (7.02 billion CNY), followed by the new district of Binhai and Wuqing district, while those in the Baodi district and Ji County are very small.HighlightsA new mathematical expression of water shortage loss was proposed in order to describe the random uncertainty and economic attributes of water resources.Gumbel copula with a new method of parameter estimation was introduced to model the joint probabilistic characteristics for water supply and water use in situations when sufficient data is unavailable.The Gumbel copula was proved to be reliable based on the RMSEs (Root mean square error) and AICs (Akaike information criterion), statistical tests and upper tail dependence tests.The potential water shortage losses for all the districts of Tianjin were predicated.

为了避免传统聚类方法和因子筛选方法的不足,本文提出一种水资源短缺风险评价耦合模型.首先引入《Science》上发表的聚类算法对水资源短缺风险进行聚类,确定建模样本的风险等级;其次引入一种标准化信息流方法检测风险因子与风险之间的因果关系,筛选水资源短缺风险敏感因子;最后采用Fisher判别分析法构建水资源短缺风险等级评价模型.针对京津唐地区水资源短缺风险评价的实例研究,表明了模型的适用性.对天津市各区县2020年的水资源短缺风险等级进行评价,研究结果表明:不考虑外调水和非传统水资源时,各区县水资源短缺风险均为高风险;考虑外调水和非传统水资源时,大部分区县水资源短缺风险等级较低,但市内六区、滨海新区、津南区、武清区和蓟县仍然处于高风险状态.加大非传统水源利用力度是降低天津市水资源短缺风险的主要途径.

Multivariate hydrological series become nonstationary under the changing environment. In this paper, a new method is proposed to study the change in the dependence structure between runoff and sediment sequences. First, a moving cut transfer entropy is proposed to detect the sudden change point in the dependence structure between runoff and sediment. Then, the Gumbel copula with a new parameter estimation method (MEE) is employed to construct the dependence structure of runoff and sediment before or after the change point for situation with small samples. The new method is applied to the annual runoff and sediment discharge series in Xiliugou River, China. It is found that the dependence structure between runoff and sediment changed abruptly in 2000. The joint probabilistic characteristics of the annual runoff and sediment discharge during the period of 1960–1999 are built based on the values of RMSEs and AICs. The joint distribution of the annual runoff and sediment discharge during the period of 2000–2016 is constructed by the Gumbel copula with MEE. Moreover, the synchronous encounter probability of runoff and sediment decreases from 0.76 to 0.63, and the asynchronous encounter probability rises from 0.24 to 0.37, after the runoff and sediment discharge series presents a significant change in the dependence structure.

The Yellow River is one of the rivers with the largest amount of sediment in the world. The amount of incoming sediment has an important impact on water resources management, sediment regulation schemes, and the construction of water conservancy projects. The Loess Plateau is the main source of sediment in the Yellow River Basin. Floods caused by extreme precipitation are the primary driving forces of soil erosion in the Loess Plateau. In this study, we constructed the extreme precipitation scenarios based on historical extreme precipitation records in the main sediment-yielding area in the middle reaches of the Yellow River. The amount of sediment yield under current land surface conditions was estimated according to the relationship between extreme precipitation and sediment yield observations in the historical period. The results showed that the extreme rainfall scenario of the study area reaches to 159.9 mm, corresponding to a recurrence period of 460 years. The corresponding annual sediment yield under the current land surface condition was range from 0.821 billion tons to 1.899 billion tons, and the median annual sediment yield is 1.355 billion tons, of which more than 91.9% of sediment yields come from the Hekouzhen to Longmen sectionand the Jinghe River basin. Therefore, even though the vegetation of the Loess Plateau has been greatly improved, and a large number of terraces and check dams have been built, the flood control and key project operation of the Yellow River still need to be prepared to deal with the large amount of sediment transport.

The evapotranspiration product of Global Land Evaporation Amsterdam Model from 1980 to 2017 was used in this study to analyze the spatial and temporal variations of ET in the headwaters of the Yellow River. The results showed that the mean annual ET in the study area was 416.8mm for the period of 1980-2017, increasing gradually from west to east. The ET presented a unimodal distribution, with lower values in January, February, November, and December, and increased significantly from May to September, with the largest ET in July and August. From 1980 to 2017, the annual ET showed a non-significant increasing trend, and the change point for ET is around 2004.