Hydrological Research Research Articles

2D hydrodynamic approach supporting evaluations of hydrological response in small watersheds: Implications for lag time estimation

The use of integrated flood models represents an approach of growing interest in the literature, being the hydrological and hydrodynamic flood processes described entirely within the two-dimensional (2D) hydrodynamic unsteady flow equations. Due to its ability in simulating complex spatial–temporal dependence of both the hydrodynamic and hydrologic responses of a catchment to rainfall events, this kind of approach paves the way to the development of novel lines of research in catchment hydrology. With particular reference to the lag time estimation, the paper focuses on three interrelated issues: 1) the use of the 2D-SWEs model to provide evidence on the variability of the lag time in small basins, 2) the description of the hydrologic response of small catchments based on characteristic times influencing the hydrodynamic response to rainfall events and 3) the representativeness of hydrodynamic-based rainfall-runoff scenarios in the description of the hydrologic response observed in real events. To that purpose, synthetic rainfall-runoff scenarios for different return periods are firstly generated to analyze the hydrologic response time in three ungauged basins. Then regressive formulas, based on ad hoc variables representing characteristic times, are introduced to interpret the variability shown by the computed lag time. Finally, the physical soundness of the proposed approach is tested against observed rainfall-runoff data in four additional basins. Despite all the simplifications introduced in the proposed approach, the regressive formulas provided a reasonably good agreement with the estimations of the lag time of the observed events, showing errors of the order of 30%. The comparison with the predictions of empirical literature formulas further confirms the potential of the proposed approach. This study represents the basis for future research on basin response to rainfall events through the use of a 2D integrated hydrodynamic-hydrologic modeling.

Reconceptualizing the hyporheic zone for nonperennial rivers and streams.

Nonperennial streams dominate global river networks and are increasing in occurrence across space and time. When surface flow ceases or the surface water dries, flow or moisture can be retained in the subsurface sediments of the hyporheic zone, supporting aquatic communities and ecosystem processes. However, hydrological and ecological definitions of the hyporheic zone have been developed in perennial rivers and emphasize the mixing of water and organisms, respectively, from both the surface stream and groundwater. The adaptation of such definitions to include both humid and dry unsaturated conditions could promote characterization of how hydrological and biogeochemical variability shape ecological communities within nonperennial hyporheic zones, advancing our understanding of both ecosystem structure and function in these habitats. To conceptualize hyporheic zones for nonperennial streams, we review how water sources and surface and subsurface structure influence hydrological and physicochemical conditions. We consider the extent of this zone and how biogeochemistry and ecology might vary with surface states. We then link these components to the composition of nonperennial stream communities. Next, we examine literature to identify priorities for hydrological and ecological research exploring nonperennial hyporheic zones. Lastly, by integrating hydrology, biogeochemistry, and ecology, we recommend a multidisciplinary conceptualization of the nonperennial hyporheic zone as the porous subsurface streambed sediments that shift between lotic, lentic, humid, and dry conditions in space and time to support aquatic-terrestrial biodiversity. As river drying increases in extent because of global change, we call for holistic, interdisciplinary research across the terrestrial and aquatic sciences to apply this conceptualization to characterize hyporheic zone structure and function across the full spectrum of hydrological states.

Comprehensive Analysis for Long-Term Hydrological Simulation by Deep Learning Techniques and Remote Sensing

Hydrological simulation plays a very important role in understanding the hydrological processes and is of great significance to flood forecasting and optimal allocation of water resources in the watershed. The development of deep learning techniques has brought new opportunities and methods for long-term hydrological simulation research at the watershed scale. Different from traditional hydrological models, the application of deep learning techniques in the hydrological field has greatly promoted the development trend of runoff prediction and provides a new paradigm for hydrological simulation. In this study, a CNN–LSTM model based on the convolutional neural network (CNN) and long short-term memory (LSTM) network, and a CNN–GRU model based on CNN and gated recurrent unit (GRN) are constructed to study the watershed hydrological processes. To compare the performance of deep learning techniques and the hydrological model, we also constructed the distributed hydrological model: Soil and Water Assessment Tool (SWAT) model based on remote sensing data. These models were applied to the Xixian Basin, and the promising results had been achieved, which verified the rationality of the method, with the majority of percent bias error (PBE) values ranging between 3.17 and 13.48, Nash–Sutcliffe efficiency (NSE) values ranging between 0.63 and 0.91, and Kling–Gupta efficiency (KGE) values ranging between 0.70 and 0.90 on a monthly scale. The results demonstrated their strong ability to learn complex hydrological processes. The results also indicated that the proposed deep learning models could provide the certain decision support for the water environment management at the watershed scale, which was of great significance to improve the hydrological disaster prediction ability and was conducive to the sustainable development of water resources.

Deep Learning for Streamflow Regionalization for Ungauged Basins: Application of Long-Short-Term-Memory Cells in Semiarid Regions

Rainfall-runoff modeling in ungauged basins continues to be a great hydrological research challenge. A novel approach is the Long-Short-Term-Memory neural network (LSTM) from the Deep Learning toolbox, which few works have addressed its use for rainfall-runoff regionalization. This work aims to discuss the application of LSTM as a regional method against traditional neural network (FFNN) and conceptual models in a practical framework with adverse conditions: reduced data availability, shallow soil catchments with semiarid climate, and monthly time step. For this, the watersheds chosen were located on State of Ceará, Northeast Brazil. For streamflow regionalization, both LSTM and FFNN were better than the hydrological model used as benchmark, however, the FFNN were quite superior. The neural network methods also showed the ability to aggregate process understanding from different watersheds as the performance of the neural networks trained with the regionalization data were better with the neural networks trained for single catchments.

Nordic hydrological frontier in the 21st century

AbstractThe 21st century has brought new challenges and opportunities and has also increased demands on the Nordic hydrological community. Our hydrological science focus and approaches need rethinking and adaptation to the changing requirements of society in response to climate change and human interventions, in search of more comprehensive and cross-disciplinary solutions. This commentary highlights new possibilities and suggests vital steps forward for the scientific discipline within Nordic hydrological research. By providing a common direction, we hope to increase awareness, accelerate progress in the hydrological community, and emphasize the importance of hydrological knowledge for serving other fields of science and society at large. We hope that our vision and the opportunities we identify will raise awareness of the scientific discipline and assist in the long-term development of the Nordic hydrological frontier in the 21st century.

NASA Power’s: an alternative rainfall data resources for hydrology research and planning activities in Bali Island, Indonesia

Rainfall data is critical for planning and research in the field of hydrology. Rainfall data must be available continuously, which means it must be recorded continuously. This recording will continue since numerous projects in the field of hydrology require continuous rainfall data. Although rainfall data are collected and recorded daily, some stations frequently have insufficient rainfall records, particularly in developing countries such as Bali, Indonesia. These issues may impair the quality of rainfall data, resulting in inaccuracies in the analysis results. To address this issue, we need a reliable source of rainfall data, one of which is NASA Power, which provides rainfall data for free. NASA Power rainfall data is then compared to observed rainfall data. The comparison of the two rainfalls is measured by a statistical parameter, namely the correlation coefficient. Based on the comparison between lowland and highland areas, the average daily rainfall from NASA Power tends to be lower than the average daily rainfall from observation stations. Meanwhile, from the correlation coefficient value (r) of the comparison of rainfall observations and NASA Power, a considerably high correlation coefficient value (> 0.7) was observed. Thus, it can be suggested that the rainfall data from NASA Power for hydrology research and planning activities in Bali Island, Indonesia.

Challenges of Hydrological Engineering Design in Degrading Permafrost Environment of Russia

The study shows that the current network of hydrometeorological observation in the permafrost zone of Russia is insufficient to provide data for the statistical approaches adopted at the state level for engineering surveys and calculations. The alternative to the financially costly and practically impossible expansion of the monitoring network is the development of hydrological research stations and the implementation of new methods for calculating streamflow characteristics based on mathematical modeling. The data of the Kolyma Water-Balance Station, the first research basin in the world in a permafrost environment (1948–1997), and the process-based hydrological model Hydrograph are applied to simulate streamflow hydrographs in remote mountainous permafrost basins. The satisfactory results confirm that mathematical modeling may substitute or replace statistical approaches in the conditions of extreme data insufficiency. The improvement of the models in a changing climate requires the renewal of historical observations at currently abandoned research stations in Russian permafrost regions. The study is important for forming the state policy in climate change adaptation and mitigation measures.

Physical Interpretation of Time‐Varying StorAge Selection Functions in a Bench‐Scale Hillslope Experiment via Geophysical Imaging of Ages of Water

AbstractUnderstanding transit times (TT) and residence times (RT) distributions of water in catchments has recently received a great deal of attention in hydrologic research since it can inform about important processes relevant to the quality of water delivered by streams and landscape resilience to anthropogenic inputs. The theory of transit time distributions (TTD) is a practical framework for understanding TT of water in natural landscapes but, due to its lumped nature, it can only hint at the possible internal processes taking place in the subsurface. While allowing for the direct observation of water movement, Electrical Resistivity Imaging (ERI) can be leveraged to better understand the internal variability of water ages within the subsurface, thus enabling the investigation of the physical processes controlling the time‐variability of TTD. In this study, we estimated time‐variable TTD of a bench‐scale bare‐soil sloping soil lysimeter through the StorAge Selection (SAS) framework, a traditional lumped‐systems method, based on sampling of output tracer concentrations, as well as through an ERI SAS one, based on spatially distributed images of water ages. We compared the ERI‐based SAS results with the output‐based estimates to discuss the viability of ERI at laboratory experiments for understanding TTD. The ERI‐derived images of the internal evolution of water ages were able to elucidate the internal mechanisms driving the time‐variability of ages of water being discharged by the system, which was characterized by a delayed discharge of younger water starting at the highest storage level and continuing throughout the water table recession.

AERA5-Asia: A Long-Term Asian Precipitation Dataset (0.1°, 1-hourly, 1951–2015, Asia) Anchoring the ERA5-Land under the Total Volume Control by APHRODITE

Abstract Accurate long-term precipitation information is critical for understanding the mechanisms behind how precipitation couples with Earth’s water fluxes, energy balances, and biogeochemical cycles across space–time scales under the changing climate. This study proposes a novel approach [daily total volume controlled merging and disaggregation algorithm (DTVCMDA)] for generating a new long-term precipitation dataset, AERA5-Asia (0.1°, 1-hourly, 1951–2015, Asia; “AERA5” is a combination of the “A” from APHRODITE and the “ERA5” from ERA5-Land), by comprehensively considering the characteristics of the high spatiotemporal resolutions and continuity of the ERA5-Land dataset and the high quality of the APHRODITE dataset. The main conclusions include, but are not limited to, the following: 1) AERA5-Asia provides time series of precipitation of sufficient resolutions, length, consistency, continuity, and quality over Asia. 2) AERA5-Asia substantially outperforms ERA5-Land and IMERG-Final in terms of both magnitudes and occurrences of precipitation events in Mainland China, especially the systematic biases. For instance, the bias of ERA5-Land, IMERG-Final, and AERA5-Asia against ground gauge-based observations in Mainland China are ∼20%, ∼11%, and ∼5%, respectively. 3) AERA5-Asia performs notably better than ERA5-Land and IMERG-Final against ground gauge-based observations in regional extreme rainfall systems (e.g., two typhoon events, Trami and Usagi). 4) AERA5-Asia should prove to be a useful precipitation dataset for addressing various key climatological and hydrological research questions that require precipitation data with longer spans and finer resolutions (0.1°, 1-hourly).

Uncertainty estimation with deep learning for rainfall–runoff modeling

Abstract. Deep learning is becoming an increasingly important way to produce accurate hydrological predictions across a wide range of spatial and temporal scales. Uncertainty estimations are critical for actionable hydrological prediction, and while standardized community benchmarks are becoming an increasingly important part of hydrological model development and research, similar tools for benchmarking uncertainty estimation are lacking. This contribution demonstrates that accurate uncertainty predictions can be obtained with deep learning. We establish an uncertainty estimation benchmarking procedure and present four deep learning baselines. Three baselines are based on mixture density networks, and one is based on Monte Carlo dropout. The results indicate that these approaches constitute strong baselines, especially the former ones. Additionally, we provide a post hoc model analysis to put forward some qualitative understanding of the resulting models. The analysis extends the notion of performance and shows that the model learns nuanced behaviors to account for different situations.

Soil thickness controls the rainfall-runoff relationship at the karst hillslope critical zone in southwest China

Hydrological processes in the critical zone are closely related to the soil–bedrock structures. However, the effect of soil thickness on the rainfall-runoff relationship on the hillslope with complex topography remains unclear. Surface runoff, lateral subsurface runoff from soil–epikarst interface, and epikarst runoff from the epikarst–bedrock interface were monitored on two adjacent plots with deep and shallow (66.0 vs. 35.4 cm) soil thicknesses from June 2019 to December 2020 in the karst region of southwest China. During the monitoring period, surface and subsurface runoff account for 20% and 37% of the total runoff in the deep-soil plot (DSP), and 3% and 43% in the shallow-soil plot (SSP). This demonstrates that runoff from the soil-epikarst system is predominant compared to the relatively small contribution of surface runoff. In the SSP, the surface topography wetness index (TWI) was highly coupled with bedrock TWI, and the bedrock TWI had a significant negative linear relationship (p < 0.01) with subsurface runoff. Moreover, isotope hydrogen-separation results showed that subsurface and epikarst runoff were dominated by pre-event water, but a higher contribution of event water was observed in the SSP than in the DSP. These findings supported the hypothesis that rainwater could infiltrate the epikarst more easily in shallow soil slopes. Rainfall and surface runoff exhibited a linear relationship in the dry season and a non-linear relationship in the rainy season, indicating the occurrence of threshold rainfall–runoff behavior. The rainfall amount threshold for surface runoff was higher in DSP (44.7 mm) than in SSP (39.5 mm), and the corresponding variation of rainfall intensity interpretation was greater (54% vs. 38%). For subsurface runoff, the rainfall amount threshold was higher in the DSP than in the SSP (91.0 vs 79.4 mm), and the corresponding variation of soil moisture interpretation was higher (56% vs. 20%). This demonstrated that runoff can be better predicted at deeper soil hillslopes by rainfall and antecedent soil moisture. Accordingly, this study emphasizes the importance of evaluating the spatial heterogeneity of soil thickness in hydrological process research.

Adaptability of MODIS Daily Cloud-Free Snow Cover 500 m Dataset over China in Hutubi River Basin Based on Snowmelt Runoff Model

Global warming affects the hydrological characteristics of the cryosphere. In arid and semi-arid regions where precipitation is scarce, glaciers and snowmelt water assume important recharge sources for downstream rivers. Therefore, the simulation of snowmelt water runoff in mountainous areas is of great significance in hydrological research. In this paper, taking the Hutubi River Basin in the Tianshan Mountains as the study area, we used the “MODIS Daily Cloud-free Snow Cover 500 m Dataset over China” (MODIS_CGF_SCE) to carry out the Snowmelt Runoff Model (SRM) simulation and evaluated the simulation accuracy. The results showed that: (1) The SRM preferably simulated the characteristics of the average daily flow variation of the Hutubi River from May to October, from 2003–2009. The monthly total runoff was maximum in July and minimum in October. Extreme precipitation events influenced the formation of flood peaks, and the interannual variation trend of total runoff from May to October was increased. (2) The mean value of the volume difference (DV) during the model validation period was 8.85%, and the coefficient of determination (R2) was 0.73. In general, the SRM underestimates the runoff of the Hutubi River, and the simulation accuracy is more accurate in the normal water period than in the high-water period. (3) By analyzing MODIS_CGF_SCE from 2003 to 2009, areas above 3200 m elevation in the Hutubi River Basin were classified as permanent snow areas, and areas below 3200 m were classified as seasonal snow areas. In October, the snow area in the Hutubi River Basin gradually increased, and the increase in snow cover in the permanent snow area was greater than that in the seasonal snow area. The snowmelt period was from March to May in the seasonal snow area and from May to early July in the permanent snow area, and the minimum snow cover was 0.7%.

A New Standardized Type Drought Indicators Based Hybrid Procedure for Strengthening Drought Monitoring System

Drought is occurring recurrently in various climatic zones around the world. Therefore, accurate and continuous drought monitoring are essential for reliable drought mitigation policies. In past research, several drought monitoring indicators have been developed. Regardless of their scopes and applicabilities, every indicator has certain amount of error regarding accurate determination of drought classes. In addition, climate change and complex features of meteorological variables also reduce the performance of each indicator. Consequently, accurate drought monitoring is a challenging task in hydrology and water management research. The objective of this research is to enhance the accuracies in drought characterization by employing multiple drought indicators, simultaneously. This article proposes a new aggregative index – the Seasonal Mixture Standardized Drought Index (SMSDI). The procedure of SMSDI is mainly based on the integration of Principle Component Analysis (PCA) and K- Component Gaussian Mixture Distribution (K-CGMD). In preliminary analysis, aggregation of three multi-scalar Standardized Drought Indices (SDIs) is made for three meteorological gauge stations of Pakistan. For comparative assessment, individual SDI has used to investigate the association and consistency with SMSDI. Outcome associated with this research shows that the SMSDI have significant correlation with individual SDIs. We conclude that instead of using individual indicator, the proposed aggregative approach enhances the scope and capacity of drought indicators for extracting reliable information related to future drought.

NeuralHydrology — A Python library for Deep Learning research in hydrology

NeuralHydrology — A Python library for Deep Learning research in hydrology

Hourly snow energy and mass balance at Atwater Study Plot, Alta, Utah

AbstractAtwater Study Plot (ASP), located in the town of Alta, UT in the Wasatch Mountains, which was established in 1939 by Monty Atwater and Ed LaChapelle, is one of the longest running snow observation sites in the Western United States. The historical observations have been primarily snow safety and avalanche forecasting focused. Beginning in 2017 the Snow Hydrology Research to Operations Laboratory (Snow HydRO Lab) at the University of Utah began regular hydrology focused snow pit observations within study plot boundaries. In 2019 the Snow HydRO Lab and the Utah Department of Transportation further supplemented ASP with an instrumentation suite for continuous snow energy and mass balance observations, including incoming and outgoing solar and terrestrial radiation. The goal of the new observations is to support process based hydrologic modelling and better understand controls on snowmelt timing and magnitude. The instrumentation installation and observations make ASP the first snow energy balance study plot in the Wasatch Mountains, and the site has since been used as a validation site for NASA's SnowEx project and is part of the NSF Critical Zone Collaborative Network. Here, we describe the study plot and present (1) the hourly instrumented dataset, which includes standard atmospheric variables (air temperature, relative humidity, and wind speed/direction), radiation balance, precipitation, and snow depth, (2) snow energy balance modelling used to calculate hourly energy exchanges and phase changes, and (3) provide directions to access the publicly available ASP datasets.

Outward Appearance or Inward Significance? On Experts' Perspectives When Studying and Solving Water Scarcity

In the first decade of the 21st century, a water harvesting approach based on contour trenches—ditches to catch runoff—from Kenya was proposed as groundwater recharge technology in a semi-arid area in Ninh Phuoc district, Vietnam. In order to modify this solutions to tackle water scarcity, hydrological conditions at the site needed to be known. For such small-scale interventions, finding the most suitable set of (cheap and quick) efforts to study local hydrology is not easy. After our own study, we explored how different experts evaluated the chosen approach. The results from this evaluation suggest that different perspective for appropriate hydrological research can be found within a group of experts. This finding—in line with anthropologically inspired science studies—suggests that integrating different perspectives from stakeholders when working on suitable solutions in real-life water scarcity situations needs to be complemented with attention for different perspectives on the underlying hydrological processes and how they are to be studied. We discuss how this notion on the multiple perspectives intrinsic to hydrological research can be fruitfully included when developing water interventions.

Realistic River Image Synthesis Using Deep Generative Adversarial Networks

In this paper, we demonstrated a practical application of realistic river image generation using deep learning. Specifically, we explored a generative adversarial network (GAN) model capable of generating high-resolution and realistic river images that can be used to support modeling and analysis in surface water estimation, river meandering, wetland loss, and other hydrological research studies. First, we have created an extensive repository of overhead river images to be used in training. Second, we incorporated the Progressive Growing GAN (PGGAN), a network architecture that iteratively trains smaller-resolution GANs to gradually build up to a very high resolution to generate high quality (i.e., 1,024 × 1,024) synthetic river imagery. With simpler GAN architectures, difficulties arose in terms of exponential increase of training time and vanishing/exploding gradient issues, which the PGGAN implementation seemed to significantly reduce. The results presented in this study show great promise in generating high-quality images and capturing the details of river structure and flow to support hydrological modeling and research.

A hydrologist's guide to open science

Abstract. Open, accessible, reusable, and reproducible hydrologic research can have a significant positive impact on the scientific community and broader society. While more individuals and organizations within the hydrology community are embracing open science practices, technical (e.g., limited coding experience), resource (e.g., open access fees), and social (e.g., fear of weaknesses being exposed or ideas being scooped) challenges remain. Furthermore, there are a growing number of constantly evolving open science tools, resources, and initiatives that can be overwhelming. These challenges and the ever-evolving nature of the open science landscape may seem insurmountable for hydrologists interested in pursuing open science. Therefore, we propose the general “Open Hydrology Principles” to guide individual and community progress toward open science for research and education and the “Open Hydrology Practical Guide” to improve the accessibility of currently available tools and approaches. We aim to inform and empower hydrologists as they transition to open, accessible, reusable, and reproducible research. We discuss the benefits as well as common open science challenges and how hydrologists can overcome them. The Open Hydrology Principles and Open Hydrology Practical Guide reflect our knowledge of the current state of open hydrology; we recognize that recommendations and suggestions will evolve and expand with emerging open science infrastructures, workflows, and research experiences. Therefore, we encourage hydrologists all over the globe to join in and help advance open science by contributing to the living version of this document and by sharing open hydrology resources in the community-supported repository (https://open-hydrology.github.io, last access: 1 February 2022).

Investigating the ability of deep learning on actual evapotranspiration estimation in the scarcely observed region

The estimation of actual evapotranspiration is a difficult issue in hydrological research, particularly in the scarcely observed region. Deep learning (DL) has been increasingly used in the field of hydrology in recent years. In this study, we investigated the ability of DL on actual evapotranspiration estimation using three sets of controlled experiments at a typical region with scarce observations, the Qinghai-Tibetan Plateau. The results suggest that the DL model can utilize a few key types of observation data to simulate actual evapotranspiration, and more in-situ observation data types did not significantly improve the accuracy of DL simulations. A multi-source DL model established by integrating data from distantly distributed stations showed a better performance than the model built separately using data at individual sites. Moreover, further analysis of climate pattern and input data correlation was conducted for the similarity of stations for multi-source learning of the DL model. Using inputs in the lead-time period can improve the simulation of daily ET by DL. Compared with traditional process-based physical methods, the DL model is more flexible to simulate actual evapotranspiration in areas with insufficient observed data such as the Qinghai-Tibetan Plateau. The results of this study highlight the potential power of DL model to improve the actual evapotranspiration estimation in the scarcely observed region.

It Takes a Village to Run a Model—The Social Practices of Hydrological Modeling

AbstractComputer models are frequently used tools in hydrological research. Many decisions related to the model set‐up and configuration have to be made before a model can be run, influencing the model results. This study is an empirical investigation of the motivations for certain modeling decisions. Fourteen modelers from three different institutes were interviewed about their modeling decisions. In total, 83 different motivations were identified. Most motivations were related to the team of the modeler and the modelers themselves, “Experience from colleagues” was the most frequently mentioned motivation. Both institutionalization and internalization were observed: a modeler can introduce a concept that subsequently becomes the teams' standard, or a modeler can internalize the default team approach. These processes depend on the experience of the modeler. For model selection, two types of motivations were identified: experience (from colleagues or the modelers themselves), and model vision (the model has assets that align with the modeling vision). Model studies are mainly driven by context, such as time constraints, colleagues, and facilities at the institute, rather than epistemic (such as aligning with the modeling vision). The role of local context in the construction of and the value assigned to models shows that models are social constructs, making model results time, and place dependent. To account for this context in the estimation of the robustness of model results, we need a diversity of opinions, perspectives, and approaches. This requires transparent modeling procedures and an explicit modeling vision for each model study.