Journal of Hydrology最新文献

Land surface models have facilitated the estimation of soil moisture over a range of spatiotemporal scales. However, limitations in model parameterization and under-representation of anthropogenic processes restrict their ability to estimate local-scale soil moisture variability, especially over irrigated areas. Assimilation of satellite-based soil moisture retrievals into land surface models can be a viable approach to overcome these constraints, specially over highly irrigated countries such as India, where such applications are rare. Additionally, large-scale validation of modeled soil moisture has been limited over India till now due to lack of a representative station network. By assimilating Soil Moisture Active Passive (SMAP)-based estimates into the state-of-the-art Indian Land Data Assimilation System (ILDAS) and combining with a new soil moisture station network of more than 200 stations, this study demonstrates improved soil moisture estimations and capture of irrigation signals over the region. The Noah-MP land surface model is forced by multiple local and global meteorological datasets and Ensemble Kalman Filter (EnKF) is used for assimilation of soil moisture. Comparison of open-loop and data assimilated soil moisture against station soil moisture data shows relative spatial mean improvement of 0.0178 in correlation and 0.0029 m³/m³ in RMSE. Further statistical comparison with in-situ data has also shown better results over most of the stations, as evident from improved correlations and reduced unbiased RMSE after assimilation. Finally, the climatology of soil moisture over the different irrigation fractions reveals that data assimilated outputs over irrigated grid cells tend to have higher soil moisture during dry winter season, demonstrating the ability to capture irrigation signals. These findings quantify the value of data assimilation in improving soil moisture estimates and the ability to capture unmodeled processes such as irrigation, which lays the science groundwork for upcoming space missions such as NASA ISRO Synthetic Aperture Radar (NISAR).

This study develops a class of robust models for flood risk mapping in highly vulnerable regions by focusing on accurately depicting extreme precipitation patterns aligned with regional climates. By implementing sophisticated hydrodynamics modeling and advanced probabilistic approaches, the present work underscores the efficacy of physical-based methodologies in the flood risk assessment. We propose a machine learning based ExGAN to address the challenge of synthesizing extreme precipitation scenarios which faithfully capture the nuances of local climatology. It is expected that through refined temporal disaggregation, the ExGAN approach exhibits exceptional proficiency in replicating a diverse spectrum of extreme precipitation patterns specific to the vulnerable region under scrutiny. Therefore, using these synthesized scenarios as inputs in a meticulously calibrated hydrological model would enable a comprehensive and detailed flood risk mapping exercise. To demonstrate the robustness of the developed mode, we perform a rigorous testing and validation within the highly susceptible Martil river basin, situated in the northern Mediterranean region of Morocco. The obtained results confirm that extending return periods would provide invaluable insights into the expanding geographical expanse of at-risk areas, clarifying the evolving landscape of vulnerability rather than merely amplifying inherent risk levels. Comparisons against the conventional Monte-Carlo sampling are also carried out in this study and the obtained results highlight significant overestimations within the latter, emphasizing the imperative need to account for diverse uncertainties beyond the basic sampling strategies within the realm of hydrodynamic modeling.

The study proposes a novel method of computing river discharge based on the maximum surface velocity recorded using a non-contact-based measurement at a singular water surface point. This location, generally, coincides with the maximum flow depth of the cross-section and accounts for the dip phenomena, where the maximum instream velocity occurs below the water surface. The method is based on information entropy theory developed by Shannon (1948) and applied to river hydraulics. In this study an alternate form of entropy is used to compute discharge as a function of the cross-sectional mean velocity, maximum velocity and shear velocity (Keulegan,1938) by minimizing the error of the state equilibrium constant, $\Phi \left(M\right)$, which is the ratio between the mean and maximum flow velocity, and that estimated using the Keulegan-based relationship. To test the accuracy of the proposed method, the maximum surface flow velocities measured at two gauging stations, each located on two different Italian rivers were studied. The estimated discharges by the proposed method were found to be comparable with the existing non-contact discharge method advocated by Moramarco et al. (2017) and, the traditional velocity-area method, using, e.g., the mean-section approach, based on the following metrics: the Nash-sutcliffe Efficiency (NSE), the coefficient of correlation and the percent bias (PBIAS). The mean velocity error emulates a Gaussian distribution for both the gauging stations and was within 95% and 5% confidance levels. Further, the entropy-based velocity profiles generated by the proposed method at the y-axis are consistent with those of the depth-based velocity profiles observed by the mechanical-current meter, thus, proving the appropriateness of the proposed discharge estimation method.

Landslide dams are composed of wide-graded materials characterized by nonuniform structures that govern breaching mechanisms. However, investigations of the failure characteristics of single-structure dams with different material compositions and inverse grading structure dams remain insufficient. In this study, a series of flume experiments are conducted to investigate the influences of noncohesive dam materials and inverse grading structures on the breaching mechanisms, hydraulic characteristics and residual dam parameters during and after landslide dam failures. The results indicate that the dam breach process is controlled by the material composition and dam structure. A coarse-grained dam remains stable with seepage, a medium-grained dam fails by headcutting and backwards erosion, and a fine-grained dam fails due to layered erosion. An inverse grading dam with coarse-grained overburden features backwards erosion or a combination of sliding and backwards erosion, while a dam with medium-grained overburden fails by headcutting and backwards erosion. The maximum erosion rate occurs at the accelerated breaching stage for single-structure dams and at the initial overtopping or accelerated breaching stage for inverse grading structure dams. Four longitudinal evolution patterns are extracted based on the breach process and erosion characteristics. In addition, the outflow discharge during dam failure can be estimated by measuring the breach width, which is defined as the straight line distance between the ends of the breach crest at the overflow face. Both the peak discharge and residual dam parameters for single-structure dams are sensitive to the median diameter of the material. These parameters of inverse grading structure dams fall within the range of values observed for dams formed by the top layer material and the bottom layer material. The initial overtopping and backwards erosion stages account for 10%–35% and 36%–66% of the total breach duration for single-structure and inverse grading structure dams, respectively. Serious errors in the prediction of breach parameters can occur when top layer materials are considered to characterize the material of inverse grading structure dams.

Impacts of climate change and human activities may lead to changes in the spatiotemporal composition of the design flood as well as its size. Previous studies mainly focused on changes in design flood size, while there has been relatively little research on changes in its regional composition. In this study, a nonstationary multi-site design flood estimation method is developed, which is useful for the design flood regional composition analysis under nonstationary conditions. Dynamic copula models are first constructed to analyze the change in the joint distribution of the nonstationary multi-site flood variables with the consideration of the nonstationarity of the marginal distribution and copula structure parameters. Then the design flood combinations in multi-site for a specified design standard are calculated by comprehensively applying the equivalent reliability method, the expectation conditional and the most-likely conditional combination strategies, which considers the future precipitation change and design lifespan length impacts on the design flood. Finally, the uncertainty of the multi-site design flood estimation caused by the model parameters uncertainty is evaluated. A case study, based on the annual maximum 7-day (AM7) flood volume in the Yichang (YC) and Cuntan (CT) sites, is conducted to illustrate this method. Results show that flood quantiles in the YC and CT sites exhibit an increasing trend as the precipitation projections will increase in the future, but the flood quantiles in the YC site are less compared to the historical period because of the huge regulation and storage effect of the Three Gorges Reservoir. In addition, the design flood combination in the CT and YC sites are calculated and the CT design floods from the expectation combination strategy are bigger than those provided by the most-likely combination strategy.

Designing city-scale Blue-Green Infrastructure (BGI) for flood risk management requires detailed and robust methods. This is due to the complex interaction of flow pathways and the need to assess cost-benefit trade-offs for various BGI options. This study aims to find a cost-effective BGI placement scheme by developing an improved approach called the Cost OptimisatioN Framework for Implementing blue-Green infrastructURE (CONFIGURE). The optimisation framework integrates a detailed hydrodynamic flood simulation model with a multi-objective optimisation algorithm (Non-dominated Sorting Genetic Algorithm II). The use of a high-resolution flood simulation model ensures the explicit representation of BGI and other land use features to simulate flow pathways and surface flood risk accurately, while the optimisation algorithm guarantees achieving the best cost-benefit trade-offs for given BGI options. The current study uses the advanced CityCAT hydrodynamic flood model to evaluate the efficiency of the optimisation framework and the impact of location and size of permeable interventions on the optimisation process and subsequent cost-benefit trade-offs. This is achieved by dividing permeable surface areas into intervention zones of varying size and quantity. Furthermore, rainstorm events with 100-year and 30-year return periods are analysed to identify any common optimal solutions for different rainfall intensities. Depending on the number of intervention locations, the automated framework reliably achieves optimal BGI implementation solutions in a fraction of the time required to find the best solutions by trialling all possible options. Designing and optimising interventions with smaller sizes but many permeable zones save a good fraction of investment. However, such a design scheme requires more computational time to find optimal options. Furthermore, the optimal spatial configuration of BGI varies with different rainstorm severities, suggesting a need for careful selection of the rainstorm return period. Based on the results, CONFIGURE shows promise in devising sustainable urban flood risk management designs.

Fractures and their geometrical patterns are usually required to analyze the mechanical and flow properties of porous media in the subsurface. Fracture characterization is therefore regarded of crucial importance for optimizing production management or achieving maximum storage capacity. In this research, we propose to invert the fracture networks under the Bayesian framework for the uncertainty quantification. In particular, the number of fractures in the modelling system is treated as unknown, leading to a trans-dimensional inverse problem, and the reversible jump Markov chain Monte Carlo algorithm is applied to sample the model space with possible model moves proposed in the sampling process. A deep learning network is further applied as a surrogate model in the sampling process for increasing the computational efficiency, instead of using the physical forward simulator. We apply the proposed methodology to estimate the spatial distribution of fracture networks based on the head measurements from the steady-state flow simulation. The prior distributions of fracture parameters such as position, orientation and length are described using the discrete fracture networks approach that is deeply rooted in stochastic modelling. Due to the high non-uniqueness, the correct spatial distribution of fracture patterns cannot be successfully recovered in this case study, even a good match between observed and simulated head data is reached. More analysis could be performed in the future with the production historical data or more informative priors.

The existing medium and long-term runoff prediction methods, which are based on data-driven and knowledge-guided methods, are associated with inherent limitations, and chaotic phenomena in runoff prediction models often leads to oscillation in the prediction error, affecting the robustness of the prediction. A knowledge-guided denoising diffusion probabilistic model (DK-RDDPM) that introduces physical theory to guide constraint quantification and obtain effective runoff uncertainty prediction results is therefore proposed in this study. The main advantage of this model is that the physical randomness in the runoff prediction process can be captured and combined with the Saint-Venant process to guide model optimization and realize more accurate medium- and long-term runoff prediction. The main contributions of this study are the establishment of a dynamic runoff probabilistic prediction model with stochastic quantification characteristics that includes the prediction uncertainty over time, and modelling of the physical constraint boundary of runoff prediction from the perspective of partial differentiation. The effectiveness of the DK-RDDPM was verified by predicting runoff in the Qijiang and Tunxi Basins in China. The results show that: 1) Encoding the physical random uncertainty operator in runoff prediction into the network of the denoising diffusion probabilistic model (DDPM) effectively captures the physically complex implicit randomness of the process, thus reducing the error that results from randomness in runoff prediction. 2) The constraint matrix that is formed using the Saint-Venant equation and the prediction matrix are layered and projected, with the fluctuation range of the constraints in each step adjusted in the optimization direction within a certain random threshold range. 3) The DK-RDDPM shows superior performance to the benchmark models, even under the influence of different noise interference factors.

The hyporheic zones (HZs) are key sites of the production of nitrous oxide (N₂O), a potent ozone-depleting greenhouse gas. Denitrification is the primary process of N₂O production in HZs, including four reduction steps (NO₃⁻→NO₂⁻→NO→N₂O→N₂). Electron competition occurs between the four reduction steps and can significantly impact the production of N₂O. However, denitrification was typically considered as simplified two step reactions for investigating the release of N₂O, neglecting the electron competition among the four-step reactions. Moreover, the N₂O production/consumption patterns are regulated by both hydraulic and biogeochemical conditions in HZs. Dynamic microbial growth cannot only mediate the biogeochemical reactions, but also change hydraulic properties spatiotemporally by bioclogging. But microbial growth is rarely considered for investigating N₂O dynamics of HZs. To assess these effects on hyporheic N₂O dynamics and source-sink function, we establish a novel numerical model of N₂O dynamics of HZs, coupling porous flow, reactive transport, electron competition, microbial growth and bioclogging. The results show that the weak electron competitiveness of N₂O reductase results in a less allocation of electrons to the N₂O reduction process, particularly in situations with limited carbon sources, thus increasing the release of N₂O into the rives. Microbial growth significantly influences N₂O release from HZs into rivers, increasing by more than two orders of magnitude on average compared to the model neglecting microbial dynamics. In contrast to the classical knowledges that HZs in coarse sediments tending to short residence time cannot act as sources of N₂O, dynamic microbial growth obviously increases the potential for N₂O release from HZs in coarse sediments to the rivers. The global Monte Carlo regional sensitivity analyses indicate that microbial biomass is the most critical factor determined the hyporheic source-sink function for N₂O, followed by carbon oxidation rate and residence time. These are significantly different from previous knowledge that the residence time and oxygen/nitrogen uptake rate are the most sensitive parameters, which may lead to misunderstanding of the key controlling factors of N₂O release from HZs. In addition, we propose a new Damköhler number (${\mathrm{Da}}_{O_2}^{\ast }$) of dissolved oxygen by multiplying the classical ${\mathrm{Da}}_{O_2}$ with a dimensionless microbial modification factor for identifying N₂O source-sink function of HZs, wit