Towards Universal Runoff Forecasting: A KAN-WLSTM Framework for Robust Multi-Basin Hydrological Modeling

Demand modelling plays a vital role in water resource management yet has rarely been critically reviewed. This paper adopts a critical realist framework for a historical analysis of demand modelling practices and their role in long-term water resource management in England and Wales from 1945 to 2010. It then focuses on recent domestic demand models in the English and Welsh private water sector. A critique of scientific realist assumptions regarding demand models is presented and the role of the current regulatory environment in encouraging a highly strategic use of demand models is discussed. Policy recommendations toward more effective modelling practices are made.

<p class="MsoNormal" style="text-align: justify; text-justify: inter-ideograph; line-height: 120%; mso-pagination: none; layout-grid-mode: char; mso-layout-grid-align: none; punctuation-wrap: simple; text-autospace: none; mso-line-break-override: restrictions; margin: 12.0pt 0cm 6.0pt 0cm;"><span lang="EN-US" style="font-size: 10.0pt; line-height: 120%; font-family: 'Times New Roman',serif; layout-grid-mode: line;">Evapotranspiration (ET) modeling plays a vital role in water resource management, agriculture, and climate adaptation. Accurate ET prediction is essential for effective irrigation planning and crop management. However, traditional methods often struggle to capture the complex relationships between environmental factors, resulting in less reliable forecasts. To address this, we implemented and optimized the Long Short-Term Memory (LSTM) network model to predict ET with improved accuracy of 98.8%, achieving a Mean Squared Error (MSE) of 0.12. Our approach incorporates SHapley Additive exPlanations (SHAP) to enhance model interpretability, offering insights into how key factors like solar radiation, wind speed, air temperature, and relative humidity impact ET predictions. The results showed that solar radiation had the highest impact on ET, followed by wind speed and air temperature. This improved understanding of key factors can help farmers and water managers make better decisions about irrigation, ensuring efficient water use and supporting sustainable agriculture. This provides a reliable and interpretable solution for ET prediction, aiding smarter irrigation strategies, improving resource efficiency, and supporting sustainable agricultural practices.</span></p>

Research has played an important role in water resource management and a consensus on research objectives would increase the efficiency of these practices. In this paper we aimed to elicit the views of attendees of the 3rd Orange River Basin Symposium regarding water-related research, by using both quantitative and qualitative responses to a questionnairesurvey, and purposeful sampling methods. Overall, research was perceived to play an important role in water resource management and there was significant agreement on which sectors are responsible for carrying out this research. Although clear strengths in water resource management in southern Africa were identified, we found that most perceived weaknessesrelated to the lack of enforcement or to human resource constraints. Despite this fact, the identified research priorities, which were aligned to those of the Water Research Commission, tended to be technical in nature and would not address these perceived weaknesses. Our recommendations were that, by incorporating previously ignored sectors into research,such as private consultants and non-governmental organisations, and addressing human capacity and enforcement issues, unique and unexplored research opportunities could improve water resource management.

Precipitation plays a critical role in water resources management, and trend changes and alterations thereof are crucial to regional or basin water security, disaster prevention, and ecological restoration under a changing environment. In order to explore the implications of precipitation variation for water resources management, taking the Wei River Basin (a transitional zone between the Guanzhong Plain and Loess Plateau) as an example, this paper proposes an index system, namely the index of precipitation alteration (IPA), to evaluate changes in precipitation and investigate their potential influence on water resources management. The system includes 17 indicators gained from observed daily rainfall, involving some structural precipitation indicators describing the precipitation patterns and some functional precipitation indicators influencing utilization of watershed water resources. Non-parametric Mann-Kendall (MK) statistical test is employed to identify the IPA trend change, and range of variability approach is used to evaluate the variation of IPA. The analysis results in Wei River Basin show that IPA varies with different spatial and temporal distributions. Overall, although the annual total precipitation declined in the study area, the frequency of extreme events was increased during 1955–2012. In the face of severe climate change patterns, it is necessary to establish the precipitation index to evaluate the change of precipitation and to provide useful information for future precipitation assessments.

Considering the future water scarcity issues, the control and management of water resources have become critically important, particularly with the rapid population growth in various countries and the rising standards of living. Agriculture, especially in developing countries, is the largest consumer of water. Therefore, the concept of virtual water gains significant importance. Virtual water refers to the hidden water embedded in agricultural products, considering the substantial amount of water required for their growth. For instance, the production of one kilogram of wheat consumes more than 1000 liters of water, termed as virtual water, as it's actually used and, in case of agricultural exports and imports, transferred from one country to another. Nations can reduce the cultivation of crops that consume a lot of water through the management of virtual water, controlling the input and output of water concerning this definition.
 Virtual water management allows countries to control their water input and output concerning their crops. This concept plays a crucial role in areas facing water scarcity. With efficient management of virtual water, demand for water in different regions (both dry and semi-arid) can be significantly reduced. The term "virtual water" indicates the amount of water used in the production of food items or commercial goods. By minimizing the production of water-intensive products, especially in water-scarce regions, the need for water consumption in those areas decreases significantly. Storing water virtually is a practical way to overcome periods of scarcity, more efficient than artificial reservoirs. The rise in competition for water resources and the increase in demand have made water management a prevalent topic. The transfer of embedded water in agricultural products is a vital component of global water management, particularly in water-scarce regions. The trade of virtual water not only saves water in importing countries but is also considered a global water conservation measure. This article aims to introduce the concept of virtual water and its role in water resource management.

Long‐term precipitation monitoring plays a vital role in water resource management and disaster prevention and mitigation. This study assesses spatial and temporal trends in seasonal and annual precipitation in Pakistan between 1960 and 2016 at an interannual scale. The Mann–Kendall (MK) test, Sen's slope (SS) estimator, and Sequential Mann–Kendall (SQMK) test were employed to assess trends. Cluster analysis and L‐moment approach were used to identify the homogenous precipitation regions. In general, increasing precipitation trends between 1960 and 2016 were evident. Results indicated increasing precipitation in winter, autumn, summer and annual scale at the rates of 0.20, 2.18, 5.16, and 10.89 mm·decade−1, respectively. In spring, the precipitation trend shows a decreasing trend at −0.67 mm·decade−1. Moreover, a significant decreasing trend occurred in winter in southern Pakistan. The overall increasing trends were more noticeable between 1960 and 1988, compared to the declining precipitation during 1989–2016. SQMK analysis indicates a clear downward trend in most regions during 1989–2016, except in autumn. Annual precipitation has increased topographically except at 500 m and 1,500 m during 1960–2016 with a significant increase of 1.37 mm·decade−1 at elevation <250 m. Results indicate a negative correlation in SS test value with seasonal and annual precipitation with elevation and a positive correlation in winter. The seasonal and annual precipitation trends exhibit increasing and decreasing trends before and after 1990, respectively, in most subregions. The notable finding based on the outcomes of this study is that the whole country observed an increasing trend during 1960–1988, followed by a decreasing trend in during 1989–2016. This decreasing tendency is particularly pronounced between 1985 and 1995, except in autumn. Agriculture production is largely reliant on precipitation in many regions. So, a detailed study of the influence of monsoon trends and large‐scale climatic variability controls over Pakistan is vital for improved water resource management in the context of global warming and rising human activity. The results will help policy makers while establishing and updating water‐related initiatives and regulations.

Reservoir dams play a pivotal role in water resource management. Accurate prediction of inflow to reservoirs significantly enhances operational performance. While standalone artificial intelligence methods have recently been frequently used to predict inflow, hybrid models have shown quite more satisfactory success. In this study, various deep learning models, including MLP, GRU, LSTM, CNN, CNN-MLP, CNN GRU, CNN-LSTM, CNN-GRU-MLP, and CNN-LSTM-MLP, were utilized to predict the monthly inflow to the Aras reservoir in the Azerbaijan-Iran region. The results were compared with the Neural Basis Expansion Analysis for Time Series Forecasting (N-BEATS) model for univariate forecasting and the NBEATSx model for multivariate forecasting using a monthly inflow time series dataset. To enhance prediction accuracy, the hyperparameters of the models were optimized. Additionally, to evaluate the impact of feature selection on model performance, five different scenarios were developed as combinations of input variables for forecasting one future time step. The evaluation metrics revealed that among the scenarios, Scenario 5 (comprising lagged inflows at months 1, 11, and 12; lagged average monthly precipitation in the upstream basin at months 1 and 12; the solar month counter; and a three-month moving average of monthly inflow) yielded the best results. Among the models, the hybrid CNN-LSTM-MLP demonstrated the highest prediction accuracy. Specifically, the performance metrics for this model and the best scenario included MAE, RMSE, PBIAS, R², KGE, and NSE, which were 8.78 m³/s, 12.95 m³/s, 1.5%, 0.89, 0.91, and 0.89, respectively. Conversely, the NBEATSx model exhibited suboptimal performance, with reduced accuracy as the number of input features increased, although the N-BEATS model performed well in univariate forecasting. This study highlights the high potential of hybrid deep learning models in accurately forecasting reservoir inflows and underscores their utility in enhancing water resource and reservoir operation management.

Conjunctive groundwater use with surface-water resources has only occasionally been considered in the hydrogeological literature and at scientific conferences (e.g. Bredehoeft and Young 1983; Sahuquillo 2002), although in terms of practical water management it represents one of the most important responses to improving drought water-supply security and for long-term climate-change adaptation, and in terms of underpinning science its design requires a refined understanding of resource interconnectivity (both naturally and perturbed by irrigation works or practices) and of aquifer salinisation processes. The aim of this essay is to provide an overview of current conjunctive use in the developing world for both irrigated agriculture and urban water-supply, and to highlight the great potential that planned ‘conjunctive management’ has as a climate-change adaptation strategy. It is primarily relevant to larger alluvial plains, which often have important rivers, large-scale irrigation systems and major aquifers in close juxtaposition—although the potential for conjunctive management can be present in a wider range of settings. There is no rigorous definition for ‘conjunctive use’; however, for the present purpose it is proposed to consider only situations where both groundwater and surface water are developed (or co-exist and can be developed) to supply a given urban area or irrigation canal-command—although not necessarily using both sources continuously over time nor providing each individual water user from either source. Adopting this rather narrow definition excludes consideration here of artificial recharge of aquifers with surface runoff or by rain-water harvesting (without direct supply from the surface-water source), use of groundwater pumping to support river baseflows (without direct supply from water wells) and catchment-scale integrated water-resources management embracing everything from flood protection to wastewater reuse (because of insufficient finance as yet to apply this in the developing world)—although it is recognised that all such techniques can play an important role in water-resources management. A key characteristic of conjunctive use is that it deploys the large natural groundwater storage associated with most aquifers to buffer the high flow variability and drought propensity of many surface watercourses (Foster et al. 2010a), and is thus capable (at varying levels of efficiency) of achieving: (1) much greater water-supply security—by taking advantage of natural aquifer storage, (2) larger net water-supply yield than would generally be possible using one source alone, (3) better timing of irrigation-water delivery—since groundwater can be rapidly deployed to compensate for any shortfall in canal-water availability at critical times for crop growth, (4) reduced environmental impact—by counteracting land water-logging and salinisation, and (5) excessive river-flow depletion or aquifer overexploitation. These benefits have been the driving force for spontaneous conjunctive use of shallow aquifers in irrigation-canal commands worldwide.

Watershed hydrologic models often possess different structures and distinct methods and require dissimilar types of inputs. As spatially-distributed data are becoming widely available, macro-scale modeling plays an increasingly important role in water resources management. However, calibration of a macro-scale grid-based model can be a challenge. The objective of this study is to improve macro-scale hydrologic modeling by joint simulation and cross-calibration of different models. A joint modeling framework was developed, which linked a grid-based hydrologic model (GHM) and the subbasin-based Soil and Water Assessment Tool (SWAT) model. Particularly, a two-step cross-calibration procedure was proposed and implemented: (1) direct calibration of the subbasin-based SWAT model using observed streamflow data; and (2) indirect calibration of the grid-based GHM through the transfer of the well-calibrated SWAT simulations to the GHM. The joint GHM-SWAT modeling framework was applied to the Red River of the North Basin (RRB). The model performance was assessed using the Nash–Sutcliffe efficiency (NSE) and percent bias (PBIAS). The results highlighted the feasibility of the proposed cross-calibration strategy in taking advantage of both model structures to analyze the spatial/temporal trends of hydrologic variables. The modeling approaches developed in this study can be applied to other basins for macro-scale climatic-hydrologic modeling.

Evaluating uncertainty estimates in distributed hydrological modeling for the Wenjing River watershed in China by GLUE, SUFI-2, and ParaSol methods

For over 30 years, IHP (International Hydrological Programme) has been actively operating as a UNESCO’s (United Nations Educational, Scientific and Cultural Organization) international scientific cooperative programme in water research, water resources management, education and capacity-building, and the only broadly-based science programme of the UN (United Nations) system in this region. By a number of initiatives and networks, the IHP has progressively carried out activities on the quantity and quality of global/regional water resources, transboundary water resources management, mitigation of water related hazards, and water education. While addressing comprehensive areas over water challenges, greater emphasis has been placed on the role of water resources management for sustainable development and with respect to the expected changes in climate and environmental conditions. WWAP (World Water Assessment Programme) and its major product WWDR (World Water Development Report) in East Asia are under the framework of IHP which supports field oriented activities on monitoring freshwater, developing case studies, enhancing national assessment capacity, and facilitating decision making processes. In light of transboundary waters in IHP, RSC (Regional Steering Committee) plays a focal role for facilitating regional cooperation in the Southeast and East Asia and Pacific States. Furthermore, ISI (International Sediment Initiative) and IFI (International Flood Initiative) have significant roles, respectively, for the management of erosion and sedimentation in line with river system or reservoir management, and for the flood management focusing on capacity building of each country in East Asia. There are other major areas of concern under UNESCO’s IHP programme in East Asia, specifically in aspects including, mitigating water conflicts on transboundary aquifers through ISARM (International Shared Aquifer Resources Management), water management of arid areas through Water and Development Information for Arid Lands- A Global Network (Asian G-WADI), and sustainable management of groundwater by UNESCO Water Chair, as well as water education through the programme of Sustainable Water Integrated Management-Educational Component.

Water resources, as the source of life, are crucial for the sustainable development of society. However, with the growth of the global population and the acceleration of industrialization, water resources are facing increasingly severe issues of scarcity and pollution. Water supply and drainage engineering plays a key role in water resource management and sustainable development. This paper explores the key technologies and methods in water resource management, with a focus on the latest developments in water resource monitoring and data collection technologies, water resource assessment and optimization methods, as well as water resource recycling and reuse technologies. Additionally, it discusses the core functions of water supply and drainage engineering in water resource supply, urban water supply and drainage system optimization, disaster response, and emergency management. Furthermore, the article analyzes the challenges and strategies for the sustainable development of water resources, presenting the current status and challenges of water resource scarcity and pollution, and exploring the balance between water resource protection and environmental impact. Finally, the paper provides a comprehensive analysis of the formulation of water resource sustainability policies and regulations, and the roles of society, economy, and technology. Through the discussion of relevant technologies and methods, this paper offers valuable insights for the optimization of future water resource management and the development of water supply and drainage engineering.

Accurate and reliable daily runoff forecasting plays a vital role in water resource management, flood warning and operational scheduling. However, runoff prediction is challenging due to its nonlinear and non‐stationary nature, influenced by climate change, topography and human activities. To improve forecasting accuracy, this study proposes a hybrid TCN‐BiLSTM model optimised by the Fruit Fly Optimization Algorithm (FOA) for daily runoff prediction in the Xijiang River basin. The model first utilises the Temporal Convolutional Network (TCN) to extract temporal features, then employs the Bidirectional Long Short‐Term Memory (BiLSTM) network to capture temporal dependencies, and finally optimises key hyperparameters of the model using the FOA to enhance overall performance. Taking the four hydrological stations in the Xijiang River basin, including WX, WZ, DHJK and GG, as examples, the model exhibits outstanding performance in both single‐step and multi‐step prediction tasks. Taking the WX station as a representative example, the model achieved an MSE, MAE, and R 2 of 0.888 × 10 6 m 3 /s, 0.530 × 10 3 m 3 /s and 0.960 on the test set, respectively. Compared with the BiLSTM model, the MSE and MAE decreased by 63.27% and 40.69%, while the R 2 increased by 7.49%. Compared with the TCN model, the MSE and MAE decreased by 59.60% and 39.71%, with an R 2 improvement of 6.31%. Relative to the TCN‐BiLSTM model, the MSE and MAE were reduced by 43.15% and 26.38%, and the R 2 increased by 3.11%. Moreover, the R 2 values for the test sets at all four stations reached 0.955 or higher, further confirming the model's stability and generalisation capability across multiple regions. The results indicate that the FOA‐TCN‐BiLSTM model demonstrates significant advantages in enhancing runoff prediction accuracy and generalisation, making it particularly suitable for practical engineering applications such as flood forecasting, water resource management, and regional hydrological risk assessment, thus holding promising application prospects.

Women’s role in water resource management is recognized, yet the implementation of methods and strategies to get beyond gender-based obstacles to women’s equal participation in water resource management related projects remain vague. Mainstream considerations on the gender aspects of development and environment focus on women as having an intrinsic relationship with the environment. Women are perceived as a natural reflection of their responsibilities for the household and the comfort and security of future generations. Contrary to mainstream environmental and political ecology research, this paper sees gender as relevant within policy and practice across multiple levels, and within institutions related to natural resource governance. Based on this, the paper looks at the sustainable development and water governance issues with the help of a specific case: the Turkey-North Cyprus Water Pipeline Project. Through broad reviews of project documentation, interviews with people who were directly involved with the project and with women’s organizations the paper draws insights on the gender aspect of the decision-making mechanisms related to water governance. The results indicate that participation by women in resource management is marginal in North Cyprus. The paper discusses that this is a reflection of a broader problem, which is gender inequality. In conclusion, one can argue that future water projects need to realize more sustainable outcomes and greater effects on gender equality in North Cyprus.

Hydrological runoff prediction plays a crucial role in water resource management and sustainable development. However, it is often constrained by the nonlinearity, strong stochasticity, and high non-stationarity of hydrological data, as well as the limited accuracy of traditional forecasting methods. Although Wasserstein Generative Adversarial Networks with Gradient Penalty (WGAN-GP) have been widely used for data augmentation to enhance predictive model training, their direct application as forecasting models remains limited. Additionally, the architectures of the generator and discriminator in WGAN-GP have not been fully optimized, and their potential in hydrological forecasting has not been thoroughly explored. Meanwhile, the strategy of jointly optimizing Variational Autoencoders (VAEs) with WGAN-GP is still in its infancy in this field. To address these challenges and promote more accurate and sustainable water resource planning, this study proposes a comprehensive forecasting model, VXWGAN-GP, which integrates Variational Autoencoders (VAEs), WGAN-GP, Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory Networks (BiLSTM), Gated Recurrent Units (GRUs), and Attention mechanisms. The VAE enhances feature representation by learning the data distribution and generating new features, which are then combined with the original features to improve predictive performance. The generator integrates GRU, BiLSTM, and Attention mechanisms: GRU captures short-term dependencies, BiLSTM captures long-term dependencies, and Attention focuses on critical time steps to generate forecasting results. The discriminator, based on CNN, evaluates the differences between the generated and real data through adversarial training, thereby optimizing the generator’s forecasting ability and achieving high-precision runoff prediction. This study conducts daily runoff prediction experiments at the Yichang, Cuntan, and Pingshan hydrological stations in the Yangtze River Basin. The results demonstrate that VXWGAN-GP significantly improves the quality of input features and enhances runoff prediction accuracy, offering a reliable tool for sustainable hydrological forecasting and water resource management. By providing more precise and robust runoff predictions, this model contributes to long-term water sustainability and resilience in hydrological systems.