Rainfall-runoff Process Research Articles

Technical note: Testing the connection between hillslope-scale runoff fluctuations and streamflow hydrographs at the outlet of large river basins

Abstract. A series of numerical experiments were conducted to test the connection between streamflow hydrographs at the outlet of large watersheds and the time series of hillslope-scale runoff yield. We used a distributed hydrological routing model that discretizes a large watershed (∼ 17 000 km2) into small hillslope units (∼ 0.1 km2) and applied distinct surface runoff time series to each unit that deliver the same volume of water into the river network. The numerical simulations show that distinct runoff delivery time series at the hillslope scale result in indistinguishable streamflow hydrographs at large scales. This limitation is imposed by space-time averaging of input flows into the river network that are draining the landscape. The results of the simulations presented in this paper show that, under very general conditions of streamflow routing (i.e., nonlinear variable velocities in space and time), the streamflow hydrographs at the outlet of basins with Horton–Strahler (H–S) order 5 or above (larger than 100 km2 in our setup) contain very little information about the temporal variability of runoff production at the hillslope scale and therefore the processes from which they originate. In addition, our results indicate that the rate of convergence to a common hydrograph shape at larger scales (above H–S order 5) is directly proportional to how different the input signals are to each other at the hillslope scale. We conclude that the ability of a hydrological model to replicate outlet hydrographs does not imply that a correct and meaningful description of small-scale rainfall–runoff processes has been provided. Furthermore, our results provide context for other studies that demonstrate how the physics of runoff generation cannot be inferred from output signals in commonly used hydrological models.

Sustainable Watershed Management: Assessing the Impact of Data Precision on Sediment Transport in the Oostanaula Creek Watershed, Tennessee

This study applies the WaTEM/SEDEM model, a watershed-scale model based on the Universal Soil Loss Equation (USLE), to enhance sustainable watershed management by identifying high-erosion-risk areas for targeted mitigation at various scales. It focuses on identifying potential errors in using statewide or worldwide land use data layers derived from remote-sensing algorithms and inaccuracies in the spatial distribution of cropland and soil types resulting in misinterpretation of sediment yields. These model limitations emphasize the need for field validation and precise input data, particularly stream data, to improve the reliability of these models. The study examines hydrologic rainfall-runoff processes in Tennessee’s 182 km2 Oostanaula catchment using various data sources, including the National Land Cover Database (NLCD), the European Space Agency WorldCover dataset (ESA), and manual field surveys. Three modeling scenarios were evaluated, with and without stream topology corrections, using the WaTEM/SEDEM model. It details the global data used, the methodology of the field survey, the simulation and validation of data, and the critical point identification. Significant discrepancies in long-term sediment transport predictions were found, depending on the land use data source. This study addresses watershed model validity and potential errors and recommendations for the use of globally available data.

Soil thickness influences the control effect of micro-topography on subsurface runoff generation in the karst hillslope critical zone

Topography plays a crucial role in predicting rainfall-runoff processes under climate change, as it affects the spatial distribution of water flow pathways. However, understanding the potential impact of high micro-topography heterogeneity on runoff generation remains challenging. We conducted high-resolution monitoring of soil moisture, instantaneous water table, and subsurface runoff in two 3D plots with different soil thicknesses (66 cm vs. 35 cm) in the karst region of southwest China. The topographic features were characterized by the soil thickness, relative elevation, and topographic wetness index obtained from a 1 × 1 m grid. The topography of the shallow-soil plot (SSP) exhibited a higher level of complexity, with a greater variability (approximately 80.0 %) in soil thickness compared to that of the deep-soil plot (DSP; approximately 61.5 %). Additionally, the SSP was prone to forming a micro-topography characterized by a combination of exposed bedrock and soil cover. In DSP, the variation in soil moisture was strongly controlled by surface topography, with elevation contributing to 52 % and 81 % of the explanation for the soil moisture at 30 cm and soil–bedrock interface, respectively. In SSP, the contribution of topography to the variation in soil moisture did not exceed 31 %. The controlling effect of topography on runoff is highlighted in SSP, with the contribution rates of surface and bedrock relative elevation reaching 79% and 97%, respectively.The weaker water storage capacity of the SSP resulted in a higher volume of subsurface runoff (161 t vs. 116 t) and frequency of runoff events (2248 vs. 1994) compared to the DSP. Accordingly, this study emphasizes that the soil–bedrock spatial heterogeneity will control the relationship between soil moisture, subsurface runoff, and terrain, which will contribute to providing a basis for flood prediction and model simulation in regions with complex topography.

Identifying components and controls of streamflow in a cold and arid headwater catchment

Streamflow is increasingly vulnerable to climate change over cold-arid regions. To understand the underlying mechanisms of streamflow changes, it is essential to distinguish streamflow components and identify their drivers due to various meltwater and rainfall runoff processes. This study combines the conceptual temperature-index Snowmelt Runoff Model (SRM) and wavelet coherence analysis to unravel the impacts of environmental factors on streamflow and its components over the semiarid-and-cold headwater catchment of the Manas River in Northwest China during 2001–2015. Our results show that SRM reproduces monthly observed streamflow with determination coefficients of > 0.91, Nash-Sutcliffe Efficiency of > 0.82, and absolute relative error of < 5 % during SRM calibration (2001–2007) and validation (2008–2015) periods. The meltwater runoff and rainfall runoff contribute to 42 % and 31 % of the total streamflow, respectively. The wavelet-based spectral analysis, partial wavelet coherency analysis, and multiple-wavelet coherence together demonstrate that total streamflow (rainfall runoff) is predominantly influenced by snow cover area (precipitation) at a scale of 32 (4–16) months. But the meltwater runoff is simultaneously controlled by precipitation, potential evapotranspiration, and normalized difference vegetation index on scales of 4–16 and > 32 months. Our study provides technical support for streamflow partitioning and controlling factors identification, and has implication to sustainable water resources management in cold and arid regions.

Flood peak estimation for the keureuto watershed using the hydrological modeling system HEC-HMS

The Keureuto watershed has been the water resource used mainly for irrigation and municipal use. Extreme rainfall along with changes in the system may trigger rainfall excess causing flood at downstream. The occurrences may impact the usability of the system that supports the water resources supply. Investigating the impact of the floods requires initial research on the rainfall-runoff process of the watershed system. Rainfall-runoff modeling is an approach widely used for estimating the discharge hydrograph of the river as the result of the processes. The HEC-HMS rainfall-runoff model tool is applied in this study to estimate the peak flood discharge for the Keureto watershed located in the North Aceh region. The data used are satellite Global Precipitation Measurement (GPM) data and Automatic Water Level Recorder (AWLR) data at the watershed outlet. The 302 km2 Keureuto Watershed is divided into 23 sub-basins for the HEC-HMS model. The SCS Curve Number is applied to the model. The simulated freturncompared to the observed flow using the Nash Sutcliffe Efficiency criteria which returns the NSE value of 0.38 for seven years simulation, based on data of year 2015 to 2021. At an event based on October 2020 to March 2021, the NSE value returns 0.52, satisfactory for the simulation data range. The highest peak of the event is captured by the model simulation in the timing and the magnitude which value lower 20% from the observed flow.

Rainfall-Runoff Process of Pallikaranai Marshland Under the Influence of Perungudi Landfill Site using HEC–HMS

The Wetland River basins are one of the critical urban watershed areas due to complex urban activities. The rainfall-runoff process in urban catchment areas is influenced by solid waste dumping, leachate generation from dumping sites, aquatic weeds, sewage generated by municipalities and town panchayats, effluents generated by industries, sand mining along the river bed and encroachment due to urban activities. Considering these complexities, this study aims to predict the stream flow in a river basin concerning a given amount of rainfall using HEC - HMS software models. The sub-basin elements are used to convert rainfall to runoff. A meteorologic model is used to assign the boundary conditions for sub-basins, which include precipitation, short/long wave radiation, and potential evapotranspiration. A time series of flow data is used as input of a model to estimate the average basin rainfall. Calibrate the model and it's required for optimization to be carried out using observed discharge. As a result, considering the three sub-basins and two reach sites, the increase in drainage area is directly proportionate to excess volume, direct run-off volume, and discharge volume, and the reliability of the model is achieved using observed data and predicted data.

Estimation of Infiltration Parameters: The Role of Pedotransfer Functions and Initial Moisture Conditions

Infiltration is one of the key components of rainfall-runoff processes at small catchment scale and is therefore a vital component of hydrological models. However, small catchments typically lack direct infiltration measurements and associated soil hydraulic characteristics, which are being substituted with soil databases and pedotransfer functions built upon these databases. We utilized two sets of pedotransfer functions Rosetta 3 and EUPTFv2 derived from two well known soil databases: UNSODA, based on US and global soil data, and EU-HYDI, based on EU soil data. Our study evaluates the impact of choosing one of these pedotransfer functions on modeled infiltration in relation to variable initial moisture content- It investigates the behavior of the simplified infiltration models of Green-Ampt and Philip under different initial conditions, in comparison with Richards’ complex solution, which is used as a benchmark. Our evaluation was conducted using soil property data available in the Czech Republic, covering ten USDA soil textures that are prevalent within the European Union. The applied initial moisture conditions were derived for specific suction pressure values that correspond to hydrological initial conditions ranging from near saturation to near the permanent wilting point. Variation in the derived parameters for the Green-Ampt and Philip models across the pedotransfer function sources results in considerable variation in the modeled infiltration. Our results imply different sensitivities of the simplified infiltration models to the initial moisture for different pedotransfer functions. The Green-Ampt and Philip infiltration models demonstrate better alignment with Richards’ complex solution with a low and consistent RMSE when using the Rosetta 3 pedotransfer function. By contrast, the EUPTFv2 model scenarios, show a clear trend of increasing RMSE towards dry initial conditions. As the European soil database is not an open data source, we only hypothesize that the distributions of soil characteristics in the European soil database differ systematically from those in UNSODA, and that this would cause different results of pedotransfer functions for different initial moisture conditions. The findings of this study are important for selecting appropriate infiltration models and, appropriate pedotransfer functions for estimating model parameters, and for assessing uncertainties connected with varying initial conditions.

Hydrologic simulation of a neotropical alpine catchment influenced by conductive topsoils in the Ecuadorian Andes

Highly conductive topsoils in neotropical high-elevation grassland-dominated ecosystems, or so-called paramos in the Andean region, influence the local rainfall-runoff processes predominated by saturation-excess overland flow as the primary source of freshwater. The Soil and Water Assessment Tool (SWAT) model has shown limitations when applied to mountainous catchments with highly conductive soils that generate surface runoff as saturation-excess overland flow. In this study, we enhanced SWAT to simulate runoff as saturation-excess overland flow and examined the hydrological responses of an intensively monitored paramo catchment in Ecuador. The model setup considered a detailed representation of the hydro-physical properties of the soils at different depths, including high infiltration and lateral flow rates in the hillslopes and restricted groundwater interactions, a characteristic of the páramo catchments. SWAT reasonably reproduced the daily discharge during dry and wet periods and the cumulative occurrence of high and low flows. The performance metrics NSE, RSR, and PBIAS values during calibration/validation period were 0.86/0.84, 0.31/0.4, and −11.2/-7.58, respectively. The runoff ratio and partitioning of the total runoff into the lateral flow and surface runoff were physically meaningful. More significantly, SWAT was able to simulate saturation-excess overland flow, which is dominant compared to infiltration excess, and it is a distinctive characteristic of páramo catchments. Nevertheless, the model showed limitations in simulating low flows.

Wavelet analysis of rainfall and application of hydrological model in a semi‐arid river basin of Rajasthan, India

AbstractRainfall–runoff modeling requires a selection of a suitable hydrologic model for the determination of an accurate quantity. A better understanding of rainfall–runoff processes is important for stream flow generation within a river basin. This study investigate the long‐term spatial–temporal trend of rainfall and streamflow using the hydrologic engineering center‐hydrologic modeling system (HEC‐HMS) model to understand the rainfall–runoff dynamics in the Banas River basin. Rainfall variability was analyzed for two periods: 1971–1995 and 1996–2020. Continuous wavelet transform was used to analyze the periodicity and the inter‐seasonal relationship in rainfall. HEC‐GeoHMS was used to generate the various inputs for the hydrologic modeling inputs for the HEC‐HMS model. Initial constant loss and ModClark transform method were used to set up the HEC‐HMS model for the study area. Calibration and validation of the model were performed using 3 years of observed data 2010–2012 and 2013–2015, respectively. Statistical model efficiency was checked using the coefficient of determination (R2), Nash–Sutcliffe efficiency, and root mean squared error to evaluate the performance of the HEC‐HMS model. The finding indicated that the R2 values are 0.80 for calibration and 0.82 for validation periods, respectively, which are in good arrangement with the measured values. This study analyzes the relationship between rainfall and runoff, which will aid in proper and efficient water resource management.

The Impact of Satellite Soil Moisture Data Assimilation on the Hydrological Modeling of SWAT in a Highly Disturbed Catchment

The potential of satellite soil moisture (SM) in improving hydrological modeling has been addressed in synthetic experiments, but it is less explored in real data cases. Here, we investigate the added value of Soil Moisture and Passive (SMAP) and Advanced Scatterometer (ASCAT) SM data to distributed hydrological modeling with the soil and water assessment tool (SWAT) in a highly human disturbed catchment (126, 486 km2) featuring a network of SM and streamflow observations. The investigation is based on the ensemble Kalman filter (EnKF) considering SM errors from satellite data using the triple collocation. The assimilation of SMAP and ASCAT SM improved the surface (0–10 cm) and rootzone (10–30 cm) SM at &gt;70% and &gt; 50% stations of the basin, respectively. However, the assimilation effects on distributed streamflow simulation of the basin are un-significant and not robust. SM assimilation improved the simulated streamflow at two upstream stations, while it deteriorated the streamflow at the remaining stations. This can be largely attributed to the poor vertical soil water coupling of SWAT, suboptimal model parameters, satellite SM data quality, humid climate, and human disturbance to rainfall-runoff processes. This study offers strong evidence of integrating satellite SM into hydrological modeling in improving SM estimation and provides implications for achieving the added value of remotely sensed SM in streamflow improvement.

Comparative Evaluation of Deep Learning Techniques in Streamflow Monthly Prediction of the Zarrine River Basin

Predicting monthly streamflow is essential for hydrological analysis and water resource management. Recent advancements in deep learning, particularly long short-term memory (LSTM) and recurrent neural networks (RNN), exhibit extraordinary efficacy in streamflow forecasting. This study employs RNN and LSTM to construct data-driven streamflow forecasting models. Sensitivity analysis, utilizing the analysis of variance (ANOVA) method, also is crucial for model refinement and identification of critical variables. This study covers monthly streamflow data from 1979 to 2014, employing five distinct model structures to ascertain the most optimal configuration. Application of the models to the Zarrine River basin in northwest Iran, a major sub-basin of Lake Urmia, demonstrates the superior accuracy of the RNN algorithm over LSTM. At the outlet of the basin, quantitative evaluations demonstrate that the RNN model outperforms the LSTM model across all model structures. The S3 model, characterized by its inclusion of all input variable values and a four-month delay, exhibits notably exceptional performance in this aspect. The accuracy measures applicable in this particular context were RMSE (22.8), R2 (0.84), and NSE (0.8). This study highlights the Zarrine River’s substantial impact on variations in Lake Urmia’s water level. Furthermore, the ANOVA method demonstrates exceptional performance in discerning the relevance of input factors. ANOVA underscores the key role of station streamflow, upstream station streamflow, and maximum temperature in influencing the model’s output. Notably, the RNN model, surpassing LSTM and traditional artificial neural network (ANN) models, excels in accurately mimicking rainfall–runoff processes. This emphasizes the potential of RNN networks to filter redundant information, distinguishing them as valuable tools in monthly streamflow forecasting.

Effects of urbanization on the water cycle in the Shiyang River basin: based on a stable isotope method

Abstract. In water-scarce arid areas, the water cycle is affected by urban development and natural river changes, and urbanization has a profound impact on the hydrological system of the basin. Through an ecohydrological observation system established in the Shiyang River basin in the inland arid zone, we studied the impact of urbanization on the water cycle of the basin using isotope methods. The results showed that urbanization significantly changed the water cycle process in the basin and accelerated the rainfall-runoff process due to the increase in urban land area, and the mean residence time (MRT) of river water showed a fluctuating downward trend from upstream to downstream and was shortest in the urban area in the middle reaches, and the MRT was mainly controlled by the landscape characteristics of the basin. In addition, our study showed that river water and groundwater isotope data were progressively enriched from upstream to downstream due to the construction of metropolitan landscape dams, which exacerbated evaporative losses of river water and also strengthened the hydraulic connection between groundwater and river water around the city. Our findings have important implications for local water resource management and urban planning and provide important insights into the hydrologic dynamics of urban areas.

A robust unstructured finite volume scheme for urban flash flood simulation: Duhok city as a case study

ABSTRACT A robust well-balanced finite volume scheme is proposed for simulating the flash floods due to rainfall-runoff process. The solver’s implementation on unstructured triangular meshes is based on a Godunov-type second-order upwind finite volume formulation, and Roe’s average scheme is used to determine the fluxes. The bed slope, friction and rainfall source terms are discretized using an upwind approach. The suggested algorithm uses the slope limiter and Hancock’s predictor-corrector method in order to secure numerical stability. The proposed model overwhelms the drawbacks of many commercial codes that use the finite difference, finite elements and finite volume in their scheme, especially in dealing with rapid change in topography. The prepared scheme solves the two-dimensional shallow water equations over a real topography of Hishkaro basin located in Duhok city in Kurdistan of Iraq where the variation of bed elevation reaches 717 m. The scheme is validated against real observed data in Duhok city provided by the Sewage directory. The data of five observation points over the study area were used to validate the code. The results showed that the max water level reaches 3.67, 2.12, 1.20, 0.67 and 0.3 m at observed points 3, 2, 4, 1 and 5, respectively.

Rainfall–runoff process and sediment yield in response to different types of terraces and their characteristics: A case study of runoff plots in Zhangjiachong watershed, China

AbstractA field experiment was conducted in the Zhangjiachong watershed to analyze the characteristics of rainfall–runoff process and sediment yield under different types of terraces. Various runoff plots were utilized, and a two serial tank model was employed to simulate runoff generation and concentration processes, taking into account vegetation interception. A soil erosion model incorporating raindrop erosion and sheet erosion was established to replicate sediment yield and concentration process. The parameters of the rainfall–runoff model and soil erosion model were calibrated and validated using observed runoff and sediment yield data. The controlling factors on runoff and sediment yield were explored and compared. The results indicate that both calibration and validation outcomes were acceptable. The earth‐banked terraces–citrus demonstrated the largest coefficients of upper and lower surface runoff, while its upper runoff threshold of the upper tank and the runoff threshold of the lower tank were the lowest, suggesting that it was more likely to produce surface and subsurface runoff. On the other hand, the stone dike terraces–crops exhibited the smallest coefficients of upper and lower surface runoff, while its upper runoff threshold of the upper tank was the highest. The lower runoff threshold of the upper tank was the same across different types of terraces. Compared to the earth‐banked terraces, the stone dike terraces have a larger infiltration coefficient in the upper tank and a smaller infiltration coefficient in the lower tank. Sensitivity analysis indicates that the most sensitive factors affecting runoff simulation results are the runoff threshold and the infiltration coefficient of the lower tank. Terraces, whether constructed with earth banks or stone dikes, are effective in reducing soil erosion. However, during storm rainfall events, these terraces may be partially destroyed, resulting in a higher sediment yield than that of the same vegetation cover on slope land. Terraces with hedgerows have a more significant effect on reducing runoff and sediment than those without hedgerows.

How do the distribution patterns of exposed roots affect the rainfall-runoff processes of sloped land under simulated multi-rainfall conditions in karst region?

Exposed root system is used as a reference of quantifying the erosion rate on slopes, but its influence mechanism of exposed root on rainfall-runoff processes during rainfall is still unclear. This study aimed to investigate the effects of exposed root distribution patterns on the rainfall-runoff processes of sloped land. Specifically, we achieved it by conducting eight multi-rainfall events under rainfall intensity of 75- and 50-mm h−1 and fixed slope gradient of 25° on the flume (1.0 m long, 0.5 m wide and 0.3 m deep) packed with yellow soils with different exposed root distribution patterns (vertical slope arrangement, parallel slope arrangement, transverse slope arrangement) on sloped land. The results revealed that during the 1st rainfall, the initial runoff time of surface, subsurface and underground runoff was shorter on parallel and vertical slope arrangement compared to transverse slope arrangement. During subsequent rainfall, surface and subsurface runoff was reduced by 5–15 times in comparison to the 1st rainfall. Furthermore, the amount of surface runoff increased in the order of parallel slope arrangement > vertical slope arrangement > transverse slope arrangement from the 1st to 4th rainfalls, while the amounts of subsurface and underground runoff decreased. On average, the amounts of surface, subsurface and underground runoff for parallel slope arrangement accounted for 61.77 %, 29.33 % and 8.90 % of the total runoff amount, respectively, whereas for vertical slope arrangement, it accounted for 57.27 %, 26.81 % and 15.92 %, and for transverse slope arrangement, it accounted for 55.78 %, 29.71 % and 14.52 % of the total runoff amount. Our findings suggested that the distribution patterns of exposed root had a significant impact on rainfall-runoff processes. Specifically, a vertical slope arrangement was conducive to rainfall infiltration, a parallel slope arrangement resulted in more surface runoff, and a transverse slope arrangement could reduce water loss, although its effect decreased with increasing rainfall frequency. In this way, the presence of exposed roots in topsoil created a rough surface that could affect the onset of flow during the initial stages of rainfall and modify the flow direction and distribution once it commenced. The results of this study would be critical in enhancing our understanding of the relationship between plant roots and hydrological processes in karst region.

Hourly rainfall-runoff modelling by combining the conceptual model with machine learning models in mostly karst Ljubljanica River catchment in Slovenia

Hydrological modelling, essential for water resources management, can be very complex in karst catchments with different climatic and geologic characteristics. In this study, three combined conceptual models incorporating the snow module with machine learning models were used for hourly rainfall-runoff modelling in the mostly karst Ljubljanica River catchment, Slovenia. Wavelet-based Extreme Learning Machine (WELM) and Wavelet-based Regression Tree (WRT) machine learning models were integrated into the conceptual CemaNeige Génie Rural à 4 paramètres Horaires (CemaNeige GR4H). In this regard, the performance of the hybrid models was compared with stand-alone conceptual and machine learning models. The stand-alone WELM and WRT models using only meteorological variables performed poorly for hourly runoff forecasting. The CemaNeige GR4H model as stand-alone model yielded good performance; however, it overestimated low flows. The hybrid CemaNeige GR4H-WELM and CemaNeige-WRT models provided better simulation results than the stand-alone models, especially regarding the extreme flows. The results of the study demonstrated that using different variables from the conceptual model, including the snow module, in the machine learning models as input data can significantly affect the performance of rainfall-runoff modelling. The hybrid modelling approach can potentially improve runoff simulation performance in karst catchments with diversified geological formations where the rainfall-runoff process is more complex.

Flood Inundation mapping and mitigation options in data-scarce region of Beledwayne town in the Wabi Shebele River Basin of Somalia

Somalia has experienced extreme flash floods in recent years across the arid regions causing tremendous loss of lives and properties. However, the flood magnitude, depth, and frequency of occurrence are not yet quantified. This is mainly due to scarce datasets in the area. . In this study, integration of observed and Climate Hazards Infrared Precipitation (CHIRP) satellite rainfall products and remote sensing raster data were used to improve hydrological model simulation outputs. The Hydrologic Engineering Center namely HEC-HMS and HEC-RAS models were used to simulate the rainfall-runoff processes and flood inundation, respectively. The land use, soil, slope and Digital Elevation map (DEM) were used to set-up the models and generate outputs. The HEC-HMS model calibration results depict that the model is able to reproduce the observed streamflow The simulated flows generated by the model predicted good agreement with the observed flow with values of 0.79, 0.74, 0.78, and 0.78 evaluated through the Nash and Sutcliffe Efficiency (NSE), Runoff Volume Error (RVE), coefficient of determination (R2), and percentage error of peak flow (PEPF), respectively. The HEC-RAS model result indicates that the maximum flood depth and velocity were obtained at the floodplain area. The peak flood at 50, 100, and 200-year return period using General Extreme Value (GEV) distribution revealed 384m3s-1, 409m3s-1, and 434m3s-1, respectively. The 100-year peak flood discharge in a specific part of the river revealed a flood depth of 7.53m. The provision of Levees as mitigation measures revealed reduction of the flood extent by 35% and suggested as possible flood protection measures for the study area.

FlowDyn: A daily streamflow prediction pipeline for dynamical deep neural network applications

This paper presents a dynamical neural network framework to understand how catchment systems respond to daily rainfall-runoff processes over time. We developed an interactive Python-based deep neural network (DNN) package called FlowDyn (presented through a JS-based web platform) to simulate and forecast daily streamflow data for >180 gauging stations across the globe. Several DNN models, including long short-term memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid network of convolutional neural network and LSTM (ConvLSTM), as well as an auto encoder (AE) model were developed and integrated into the FlowDyn pipeline to analyze and forecast sequential daily streamflow values that are embedded within a web-based application for demonstration and visualization. Inputs were gathered from different web services, including the catchment attributes and meteorology for large-sample studies (CAMELS), the national climatic data center (NCDC), and the global runoff data center (GRDC). DNN configurations were trained and tested with an average accuracy rating of 0.83 across 183 river basins globally. FlowDyn simulation and performance demonstrated that different DNN models were able to learn both regionally consistent and location-specific hydrological behaviors. Through the findings of this paper, we advocate the merit of applying FlowDyn package in the field of daily rainfall-runoff prediction at both local and global scales.

Changes in Stormwater Quality and Heavy Metals Content along the Rainfall–Runoff Process in an Urban Catchment

Stormwater quality in an urban watershed can be influenced by several factors, including land use patterns, atmospheric deposition, and human activities. The objective of this study was to investigate spatial and temporal changes in stormwater quality and heavy metal content during the rainfall–runoff in an urban sub-catchment (30 ha) in the town of Olsztyn (NE Poland). Samples were collected from six locations along the rainfall–runoff pathway, including the following direct rainfall and runoff locations: roof runoff, surface runoff, storm collector, and the river. Parameters such as pH, specific conductivity, fluorescent dissolved organic matter (fDOM), total dissolved solids (TDS), and turbidity were measured in situ, while samples were analyzed for heavy metal content (Cu, Cr, Fe, Ni, Zn, and Pb) in the lab (ICP-OES). The results showed significant changes in water quality along the runoff. The highest concentrations of heavy metals were found in samples from a stormwater collector and surface runoff, particularly in winter and spring, due to the increased deposition of air pollutants and salt washout from roads. This study highlights the importance of monitoring stormwater quality and heavy metals in urban watersheds in terms of impacts on the river ecosystem as a recipient of stormwater. Solutions such as green infrastructure and stormwater management are proposed to mitigate the impacts of urbanization on water quality and protect the aquatic environment.

Development of the Synthetic Unit Hydrograph Tool – SUnHyT

Unit hydrographs (UH) are widely used in scientific research and engineering projects to simulate rainfall-runoff processes. There are four main approaches for calculating UH: the traditional, the conceptual, the probabilistic, and the geomorphological approaches. Most software designed to facilitate the estimation of UH is usually based on only one UH approach, limiting its applicability for scientific hypotheses testing. This paper presents the Synthetic Unit Hydrograph Tool (SUnHyT), which provides nine different UH models from the four main approaches used in UH applications. SUnHyT is an open-source application that can be used intuitively through a graphical user interface. We tested the model in a case study that highlights the need for alternative approaches of UH when the traditional approach does not perform well. SUnHyT allows the estimation of design hydrographs in gauged and ungauged catchments and can be useful for hydrologists, water managers and decision-makers.