Catchment Descriptors Research Articles

Gauged and historical abrupt wave front floods (‘walls of water’) in Pennine rivers, northern England

AbstractExtremely rapid rates of rise in level and discharge in a subset of flash floods (‘abrupt wave front floods’, AWF) are separate hazards from peak level. Such flood events are investigated for Pennine catchments in northern England using both gauged and historical information. Gauged level and flow digital records at 15‐min intervals provide recent data. Historical information for 122 AWF events is extracted from a chronology of flash floods for Britain. Historical AWF events are mapped and found to occur on every major Pennine catchment; catchment descriptors are derived as a basis for assessing catchment vulnerability. We discuss the disputed origin of AWF. Using gauged data, we contrast the rising limb of AWF and ‘normal’ floods. We investigate time series of historical AWF, noting a puzzling peak in the late 19th century. Current rainfall and river monitoring does not provide a reliable basis for understanding AWF processes or for operational response and we suggest improvements. Similarly, current models for design flood estimation and forecasting do not generate the observed rapid increase in level in AWF floods.

FOCA: a new quality-controlled database of floods and catchment descriptors in Italy

Abstract. Here we present FOCA (Italian FlOod and Catchment Atlas), the first systematic collection of data on Italian river catchments for which historical discharge time series are available. Hydrometric information, including the annual maximum peak discharge and average daily annual maximum discharge, is complemented by several geomorphological, climatological, extreme rainfall, land-cover and soil-related catchment attributes. All hydrological information derives from the most recently released datasets of discharge and rainfall measurements. To enhance the reproducibility and transferability of the analysis, this paper provides a description of all the raw data and the algorithms used to build the basin attribute dataset. We also describe the approaches adopted to solve problems encountered during the digital elevation model elaboration in areas characterized by a complex morphology. Details about the data quality-control procedure developed to detect and correct errors are also reported. One of the main novelties of FOCA with respect to other national-scale datasets is the inclusion of a rich set of geomorphological attributes and extreme rainfall features for a large set of basins covering a wide range of elevations and areas. Using this first nationwide data collection (available at https://doi.org/10.5281/zenodo.10446258, Claps et al., 2023), a wide range of environmental applications, with a particular focus on flood studies, can be undertaken within the Italian territory.

Nonlinearities in phytoplankton groups across temperate high mountain lakes

Abstract High mountain lakes are increasingly recognized as sentinel ecosystems of global change. Monitoring phytoplankton changes or reconstructing their composition from sedimentary records can help identify systemic changes in these lakes and their catchments. This study aimed to evaluate the distribution of the major phytoplankton groups in high mountain lakes across environmental gradients and identify tipping points in relative dominance. The phytoplankton groups were estimated using pigment‐based chemotaxonomy in 79 lakes in the Pyrenees selected to cover the bedrock and elevation gradients. Fifty‐four environment variables were considered, including in‐lake and catchment descriptors. Redundancy analyses showed that in‐lake descriptors override the explicative capacity of landscape variables. Generalized additive models and multivariate regression trees showed that water hardness, trophic state and food‐web descriptors were, in this order, the most influential factors determining phytoplankton group dominance. Calcium concentration of about 200 μeq L−1 defined the threshold between soft waters—with chrysophytes and chlorophytes showing a higher affinity for them—and harder waters that favour diatoms and cyanobacteria. Across the trophic gradient, there was a threshold at ~5 μg L−1 of total phosphorus (TP), chrysophytes being dominant below that TP value and cryptophytes above. The dominance of chlorophytes and cryptophytes increased with the density of macrozooplankton. Chrysophytes were significantly lower and diatoms higher in lakes with fish. Synthesis. The relative abundance of phytoplankton groups in temperate high mountain lakes responds in a nonlinear way to the hardness of the water in the range 20–1195 Ca2+ μeq L−1 and the trophic state in the range 0.94–19 TP μg L −1. The thresholds across water hardness and trophic state gradients coincide with studies based on other organisms, pointing to a robust typology for mountain lakes that should be considered when selecting global change sentinel lakes and anticipating abrupt transitions across these thresholds.

Analysis of scaling relationships for flood parameters and peak discharge estimation in a tropical region

Abstract Relationships between peak discharges and catchment size (e.g., flood scaling) in a catchment have the potential to support new river flood forecasting approaches but have not been tested in tropical regions. This study determined flood scaling relationships between peak discharge and nested drainage areas in the La Sierra catchment (Mexico). A statistical power law equation was applied to selected rainfall–runoff events that occurred between 2012 and 2015. Variations in flood scaling parameters were determined in relation to catchment descriptors and processes for peak downstream discharge estimation. Similar to studies in humid temperate regions, the results reveal the existence of log-linear relationships between the intercept (α) and exponent (θ) parameter values and the log–log power–law relationships between (α) and the peak discharge observed from the smallest headwater catchments. The flood parameter values obtained were then factored into the scaling equation (QP = αAθ) and successfully predicted downstream flood peaks, especially highly recurrent flood events. The findings contribute to a better understanding of the nature of flood wave generation and support the development of new flood forecasting approaches in unregulated catchments suitable for non-stationarity in hydrological processes with climate change.

Towards robust seasonal streamflow forecasts in mountainous catchments: impact of calibration metric selection in hydrological modeling

Abstract. Dynamical (i.e., model-based) methods are widely used by forecasting centers to generate seasonal streamflow forecasts, building upon process-based hydrological models that require parameter specification (i.e., calibration). Here, we investigate the extent to which the choice of calibration objective function affects the quality of seasonal (spring–summer) streamflow hindcasts produced with the traditional ensemble streamflow prediction (ESP) method and explore connections between hindcast skill and hydrological consistency – measured in terms of biases in hydrological signatures – obtained from the model parameter sets. To this end, we calibrate three popular conceptual rainfall-runoff models (GR4J, TUW, and Sacramento) using 12 different objective functions, including seasonal metrics that emphasize errors during the snowmelt period, and produce hindcasts for five initialization times over a 33-year period (April 1987–March 2020) in 22 mountain catchments that span diverse hydroclimatic conditions along the semiarid Andes Cordillera (28–37∘ S). The results show that the choice of calibration metric becomes relevant as the winter (snow accumulation) season begins (i.e., 1 July), enhancing inter-basin differences in hindcast skill as initializations approach the beginning of the snowmelt season (i.e., 1 September). The comparison of seasonal hindcasts shows that the hydrological consistency – quantified here through biases in streamflow signatures – obtained with some calibration metrics (e.g., Split KGE (Kling–Gupta efficiency), which gives equal weight to each water year in the calibration time series) does not ensure satisfactory seasonal ESP forecasts and that the metrics that provide skillful ESP forecasts (e.g., VE-Sep, which quantifies seasonal volume errors) do not necessarily yield hydrologically consistent model simulations. Among the options explored here, an objective function that combines the Kling–Gupta efficiency (KGE) and the Nash–Sutcliffe efficiency (NSE) with flows in log space provides the best compromise between hydrologically consistent simulations and hindcast performance. Finally, the choice of calibration metric generally affects the magnitude, rather than the sign, of correlations between hindcast quality attributes and catchment descriptors, the baseflow index and interannual runoff variability being the best predictors of forecast skill. Overall, this study highlights the need for careful parameter estimation strategies in the forecasting production chain to generate skillful forecasts from hydrologically consistent simulations and draw robust conclusions on streamflow predictability.

River reach-level machine learning estimation of nutrient concentrations in Great Britain

Nitrogen (N) and phosphorus (P) are essential nutrients necessary for plant growth and support life in aquatic ecosystems. However, excessive N and P can lead to algal blooms that deplete oxygen and lead to fish death and the release of toxins that are harmful to humans. Estimates of N and P levels in rivers are typically calculated at station or grid (>1 km) scale; therefore, it is difficult to visualise the evolution of water quality as water travels downstream. Using a high-resolution reach-scale river network and associating each reach with land cover fractions and catchment descriptors, we trained random forest models on aggregated data (2010–2020) from the Environmental Agency Open Water Quality Data Archive for 2,343 stations to predict long-term nitrate and orthophosphate concentrations at each river reach in Great Britain (GB). We separated the model training and predictions for different seasons to investigate the potential difference in feature importance. Our model predicted concentrations with an average testing coefficient of determination (R2) of 0.71 for nitrate and 0.58 for orthophosphate using 5-fold cross-validation. Our model showed slightly better performance for higher Strahler stream orders, highlighting the challenges of making predictions in small streams. Our results revealed that arable and horticultural land use is the strongest and most reliable predictor for nitrate, while floodplain extents and standard percentage runoff are stronger predictors for orthophosphate. Nationally, higher orthophosphate concentrations were observed in urbanised areas. This study shows how combining a river network model with machine learning can easily provide a river network understanding of the spatial distribution of water quality levels.

Driving Factors of the Hydrological Response of a Tropical Watershed: The Ankavia River Basin in Madagascar

Understanding the hydrological behavior of watersheds (WS) and their driving factors is crucial for sustainable water resources management. However, at large scales, this task remains challenging due to the spatial heterogeneity in landscapes, topography and morphology (T), land cover (LC), geology (G), and soil properties (S). In this context, the aim of this study was to identify the key factors that influence the hydrological signatures of four watersheds: Ankavia (WS1: 55% forest cover), Ankaviabe (WS2: 77% forest cover), Sahafihitry (WS3: 41% forest cover), and Antsahovy (WS4: 48% forest cover), over a 10-month study period. These catchments are located within the SAVA region of northeastern Madagascar and have a humid tropical climate. We investigated the relationship between selected catchment descriptors and hydrological signatures by using a Pearson coefficient-based correlation matrix. More specifically, catchment descriptors (extracted from T, LC, G, and S) were correlated with the following hydrological signatures: base flow index (BFI), mean runoff coefficient (rc), mean peak flow (Qp), mean runoff event time scales (ts), high flows (Q5), low flows (Q95), and mean discharge (q_mean). The analysis revealed that land cover, soil properties, and geology seem to be the best predictors for BFI and Q95, while soil properties mainly govern rc, Qp, Q5, ts, and q_mean. These findings provide valuable insights into the key drivers of hydrological behavior that can inform water resource management strategies. In particular, WS2 exhibits better flood buffering capacity but also experiences lower base flows in the dry season, potentially due to higher evapotranspiration. Conversely, WS3 and WS4 (and to a lesser extent WS1) have lower flood buffering capacity, but these watersheds encounter less pronounced low flows in the dry season due to higher BFIs, possibly attributable to lower evapotranspiration rates. The results underscore the importance of responsible land use practices and conservation efforts, which are essential for the sustainable development of the region. By incorporating these practices into water management strategies, we can help ensure a more stable and reliable water supply for communities and ecosystems within the region.

Exploring controls on rainfall–runoff events: spatial dynamics of event runoff coefficients in Iran

ABSTRACT This study investigates the controls that influence the spatial variability of event runoff coefficients (RCs) based on the comparative hydrology concept. We analysed 18 468 storm events in 963 Iranian catchments ranging in size from 101 to 8467 km2. The mean RCs vary spatially due to long-term hydro-climate conditions and other catchment descriptors (CDs). The controls on spatial variability of RCs are investigated using a correlation analysis of the statistical moments of the RCs and CDs. The results indicate (1) catchment elevation is the most important control on RCs at the country scale, and hydro-climatic attributes such as mean annual precipitation, ratio of actual evaporation to precipitation, and the ratio of potential evaporation to precipitation are other important controls on RCs; and (2) at the regional scale, catchment area is a major control on RCs, although it is not as important as hydro-climatic controls at the national scale.

Estimation of stream water components and residence time in a permafrost catchment in the central Tibetan Plateau using long-term water stable isotopic data

Abstract. Global warming has significantly impacted the hydrological processes and ecological environment in permafrost regions. Mean residence time (MRT) is a fundamental catchment descriptor that provides hydrological information regarding storage, flow pathways, and water source within a particular catchment. However, water stable isotopes and MRT have rarely been investigated due to limited data collection in the high-altitude permafrost regions. This study uses the long-term stable isotopic observations to identify runoff components and applied the sine-wave exponential model to estimate water MRT in a high-altitude permafrost catchment (5300 m a.s.l.) in the central Tibetan Plateau (TP). We found that the isotope composition in precipitation, stream, and supra-permafrost water exhibited obvious seasonal variability. The freeze–thaw process of the permafrost active layer and direct input of precipitation significantly modified the stable isotope compositions in supra-permafrost and stream water. The hydrograph separation revealed that precipitation and supra-permafrost water accounted for 35 ± 2 % and 65 ± 2 % of the total discharge of stream water, respectively. MRT for stream and supra-permafrost water was estimated at 100 and 255 d, respectively. Such shorter MRTs of supra-permafrost and stream water (compared to the non-permafrost catchments) might reflect the unique characteristics of the hydrological process in permafrost catchments. Moreover, the MRT of supra-permafrost water was more sensitive to environmental change than that of stream water. Climate and vegetation factors affected the MRT of stream and supra-permafrost water mainly by changing the thickness of the permafrost active layer. Our results suggest that climate warming might retard the rate of water cycle in permafrost regions. Overall, our study expands our understanding of hydrological processes in high-altitude permafrost catchments under global warming.

Design flood estimation using extreme Gradient Boosting-based on Bayesian optimization

• Proposed XGB-based ML model for RFFA and flood estimation. • CAMELs data for 671 catchments from the USA were used to test the efficacy of the approach. • Multiple Linear Regression methods and Artificial Neural Networks were used to compare the output of the eXtreme Gradient Boost-based ML model. • The importance of each catchment feature is characterized using Gini Impurity, Permutation, and Dropout Loss Feature Ranking methods. Regional Flood Frequency Analysis (RFFA) is one of the widely used approaches for estimating design floods in the ungauged basins. We developed an eXtreme Gradient Boost (XGB) machine learning model for RFFA and flood estimation. Our approach relies on developing a regression model between flood quantiles and the commonly available catchment descriptors. We used CAMELs data for 671 catchments from the USA to test the approach's efficacy. The results were compared with the traditional Multiple Linear Regression methods and Artificial Neural Networks. Results revealed that the XGB-based approach estimated design flood with the highest accuracy during training and validation with minor mean absolute error, root mean square error values, and percentage bias ranging from −10 to + 10. The importance of each catchment feature is visualized by three different approaches Gini Impurity, Permutation, and Dropout Loss Feature Ranking. We observed that the most dominating variables are rainfall intensity, slope, snow fraction, soil porosity, and temperature. It is observed that the importance of these variables is a function of the hydroclimatic regions and varies with space. In contrast, mean annual areal potential evapotranspiration, mean annual rainfall, fraction forest area, and soil conductivity have low significance in estimating design flood for an ungauged catchment. Indeed, the proposed XGB-based approach has broader applicability and replicability.

Exploring the dependence between discharge indices and catchment descriptors for suspended sediment load in Indian Rivers

ABSTRACT In this paper, we have explored the dependence of discharge indices (and their return period) on various catchment descriptors, such as average slope, basin relief, drainage area, maximum elevation, and mean elevation. Discharge indices, viz., effective discharge, functional-equivalent discharge, and fraction-load discharge, are considered in this study. These indices characterize long-term sediment transport through stream networks and are generally adopted as design flows for restoration of streams and in geomorphic, as well as in eco-hydrologic studies. These indices (and their return period) were determined based on a continuous probability distribution (e.g. log-normal and Gamma)-based Magnitude-Frequency Analysis approach. These indices and their return period were found to have a negative correlation with all these catchment descriptors except the drainage area. This is because there is a reduction in the magnitude of discharge indices with the rise in average slope, basin relief, maximum elevation, and mean elevation, which indicates that a lesser magnitude of flows/discharges is effective in carrying a considerable amount of sediments over time as the transport of sediments is assisted by high slopes of upstream catchments, viz., headwater catchments. These correlated descriptors can be useful in the determination of indices (and their return period) for ungauged catchments.

Spatio‐temporal discretization uncertainty of distributed hydrological models

AbstractQuantifying the uncertainty linked to the degree to which the spatio‐temporal variability of the catchment descriptors (CDs), and consequently calibration parameters (CPs), represented in the distributed hydrology models and its impacts on the simulation of flooding events is the main objective of this paper. Here, we introduce a methodology based on ensemble approach principles to characterize the uncertainties of spatio‐temporal variations. We use two distributed hydrological models (water balance simulation model and Hydrotel) and six catchments with different sizes and characteristics, located in southern Quebec, to address this objective. We calibrate the models across four spatial (100, 250, 500 and 1000 m2) and two temporal (3 and 24 h) resolutions. Afterwards, all combinations of CDs‐CPs pairs are fed to the hydrological models to create an ensemble of simulations for characterizing the uncertainty related to the spatial resolution of the modelling, for each catchment. The catchments are further grouped into large (>1000 km2), medium (between 500 and 1000 km2) and small (<500 km2) to examine multiple hypotheses. The ensemble approach shows a significant degree of uncertainty (over 100% error for estimation of extreme streamflow) linked to the spatial discretization of the modelling. Regarding the role of CDs, results show that first, there is no meaningful link between the uncertainty of the spatial discretization and catchment size, as spatio‐temporal discretization uncertainty can be seen across different catchment sizes. Second, the temporal scale plays only a minor role in determining the uncertainty related to spatial discretization. Third, the more physically representative a model is, the more sensitive it is to changes in spatial resolution. Finally, the uncertainty related to model parameters is larger than that of CDs for most of the catchments. Yet, there are exceptions for which a change in spatio‐temporal resolution can alter the distribution of state and flux variables, change the hydrologic response of the catchments and cause large uncertainties.

Approximate Bayesian inference for analysis of spatiotemporal flood frequency data

Extreme floods cause casualties and widespread damage to property and vital civil infrastructure. Predictions of extreme floods, within gauged and ungauged catchments, is crucial to mitigate these disasters. In this paper a Bayesian framework is proposed for predicting extreme floods, using the generalized extreme-value (GEV) distribution. A major methodological challenge is to find a suitable parametrization for the GEV distribution when multiple covariates and/or latent spatial effects are involved and a time trend is present. Other challenges involve balancing model complexity and parsimony, using an appropriate model selection procedure and making inference based on a reliable and computationally efficient approach. We here propose a latent Gaussian modeling framework with a novel multivariate link function designed to separate the interpretation of the parameters at the latent level and to avoid unreasonable estimates of the shape and time trend parameters. Structured additive regression models, which include catchment descriptors as covariates and spatially correlated model components, are proposed for the four parameters at the latent level. To achieve computational efficiency with large datasets and richly parametrized models, we exploit a highly accurate and fast approximate Bayesian inference approach which can also be used to efficiently select models separately for each of the four regression models at the latent level. We applied our proposed methodology to annual peak river flow data from 554 catchments across the United Kingdom. The framework performed well in terms of flood predictions for both ungauged catchments and future observations at gauged catchments. The results show that the spatial model components for the transformed location and scale parameters as well as the time trend are all important, and none of these should be ignored. Posterior estimates of the time trend parameters correspond to an average increase of about 1.5% per decade with range 0.1% to 2.8% and reveal a spatial structure across the United Kingdom. When the interest lies in estimating return levels for spatial aggregates, we further develop a novel copula-based postprocessing approach of posterior predictive samples in order to mitigate the effect of the conditional independence assumption at the data level, and we demonstrate that our approach indeed provides accurate results.

Runoff Modeling in Ungauged Catchments Using Machine Learning Algorithm-Based Model Parameters Regionalization Methodology

Model parameters estimation is a pivotal issue for runoff modeling in ungauged catchments. The nonlinear relationship between model parameters and catchment descriptors is a major obstacle for parameter regionalization, which is the most widely used approach. Runoff modeling was studied in 38 catchments located in the Yellow–Huai–Hai River Basin (YHHRB). The values of the Nash–Sutcliffe efficiency coefficient (NSE), coefficient of determination (R2), and percent bias (PBIAS) indicated the acceptable performance of the soil and water assessment tool (SWAT) model in the YHHRB. Nine descriptors belonging to the categories of climate, soil, vegetation, and topography were used to express the catchment characteristics related to the hydrological processes. The quantitative relationships between the parameters of the SWAT model and the catchment descriptors were analyzed by six regression-based models, including linear regression (LR) equations, support vector regression (SVR), random forest (RF), k-nearest neighbor (kNN), decision tree (DT), and radial basis function (RBF). Each of the 38 catchments was assumed to be an ungauged catchment in turn. Then, the parameters in each target catchment were estimated by the constructed regression models based on the remaining 37 donor catchments. Furthermore, the similarity-based regionalization scheme was used for comparison with the regression-based approach. The results indicated that the runoff with the highest accuracy was modeled by the SVR-based scheme in ungauged catchments. Compared with the traditional LR-based approach, the accuracy of the runoff modeling in ungauged catchments was improved by the machine learning algorithms because of the outstanding capability to deal with nonlinear relationships. The performances of different approaches were similar in humid regions, while the advantages of the machine learning techniques were more evident in arid regions. When the study area contained nested catchments, the best result was calculated with the similarity-based parameter regionalization scheme because of the high catchment density and short spatial distance. The new findings could improve flood forecasting and water resources planning in regions that lack observed data.

A modified approach to determine suspended sediment transport effectiveness in Indian rivers

Transport Effectiveness (TE) of a discharge is defined as the product of the frequency of discharge and corresponding sediment transport rate. Maximization of TE function helps in finding the “effective discharge (Qe)” which is responsible for the transfer of the majority of sediments over a prolonged period. This approach to determine Qe is famously known as Magnitude-Frequency Analysis (MFA). Conventionally, MFA involves the construction of the TE function by assuming (i) a location-specific probability density function (PDF) for discharge and (ii) power-law relationship between discharge and sediment transport rate. There have been attempts in the past to derive expressions for Qe by assuming positively skewed PDFs for discharge. The present study proposes a modified approach to estimate TE-based Qe for general discharge distribution datasets which alleviates the limitation of assuming a location-specific PDF. The approach involves the transformation of discharge data through Box-Cox transformation and subsequently performing MFA to determine Qe. An expression for the analytical solution of TE-based Qe is derived for the proposed framework of MFA. The robustness of the approach is established through a simulation study by generating discharge data using lognormal, Gamma, and log Pearson type III distributions. In this study, the influence of the hysteresis effect was investigated for total as well as seasonal datasets according to the considerations of the rising and falling stages for 14 stream gauges in South Indian Rivers. The analysis is performed by using the suspended sediment load and discharge data whereas the bedload has been excluded. Subsequently, the estimates of Qe based on the proposed approach (QePA) were computed considering various sediment rating curves fitted for total, seasonal, stage and season-stage based datasets. Finally, the variation of QePA with various catchment descriptors was examined.

Prediction of flood quantiles at ungauged catchments for the contiguous USA using Artificial Neural Networks

Abstract In this study, we utilise Artificial Neural Network (ANN) models to estimate the 100- and 1500-year return levels for around 900,000 ungauged catchments in the contiguous USA. The models were trained and validated using 4,079 gauges and several selected catchment descriptors out of a total of 25 available. The study area was split into 15 regions, which represent major watersheds. ANN models were developed for each region and evaluated by calculating several performance metrics such as root-mean-squared error (RMSE), coefficient of determination (R2) and absolute percent error. The availability of a large dataset of gauges made it possible to test different model architectures and assess the regional performance of the models. The results indicate that ANN models with only one hidden layer are sufficient to describe the relationship between flood quantiles and catchment descriptors. The regional performance depends on climate type as models perform worse in arid and humid continental climates. Overall, the study suggests that ANN models are particularly applicable for predicting ungauged flood quantiles across a large geographic area. The paper presents recommendations about future application of ANN in regional flood frequency analysis.

Exploring the Application of Flood Scaling Property in Hydrological Model Calibration

AbstractModel calibration has always been one major challenge in the hydrological community. Flood scaling properties (FS) are often used to estimate the flood quantiles for data-scarce catchments based on the statistical relationship between flood peak and contributing areas. This paper investigates the potential of applying FS and multivariate flood scaling properties [multiple linear regression (MLR)] as constraints in model calibration. Based on the assumption that the scaling property of flood exists in four study catchments in northern China, eight calibration scenarios are designed with adopting different combinations of traditional indicators and FS or MLR as objective functions. The performance of the proposed method is verified by employing a distributed hydrological model, namely, the Soil and Water Assessment Tool (SWAT) model. The results indicate that reasonable performance could be obtained in FS with fewer requirements of observed streamflow data, exhibiting better simulation of flood peaks than the Nash–Sutcliffe efficiency coefficient calibration scenario. The observed streamflow data or regional flood information are required in the MLR calibration scenario to identify the dominant catchment descriptors, and MLR achieves better performance on catchment interior points, especially for the events with uneven distribution of rainfall. On account of the improved performance on hydrographs and flood frequency curve at the watershed outlet, adopting the statistical indicators and flood scaling property simultaneously as model constraints is suggested. The proposed methodology enhances the physical connection of flood peak among subbasins and considers watershed actual conditions and climatic characteristics for each flood event, facilitating a new calibration approach for both gauged catchments and data-scarce catchments.Significance StatementThis paper proposes a new hydrological model calibration strategy that explores the potential of applying flood scaling properties as constraints. The proposed method effectively captures flood peaks with fewer requirements of observed streamflow time series data, providing a new alternative method in hydrological model calibration for ungauged watersheds. For gauged watersheds, adopting flood scaling properties as model constraints could make the hydrological model calibration more physically based and improve the performance at catchment interior points. We encourage this novel method to be adopted in model calibration for both gauged and data-scarce watersheds.

Assessment of dimension-reduction and grouping methods for catchment response time estimation in Hungary

Study region61 catchments located in Hungary, with drainage areas from 8.74 to 810 km2. Study focusMany engineering tasks require the estimation of the catchment response time (Tr). The most frequently used Tr parameters are the time of concentration and the lag time. At ungauged catchments, they are usually estimated by empirical equations that relate Tr to catchment characteristics. This paper provides a comparative study of three dimension-reduction techniques and seven clustering methods for fitting empirical equations to the observed values of Tr. 60 catchment descriptors were calculated for each catchment, then three subsets with 1–3 descriptors were extracted from the entire parameter set during the dimension-reduction analysis. Two and four catchment groups were created during a cluster analysis, by re-calibrating the three equations that resulted from the dimension-reduction analysis. New hydrological insights for the region under studyIt is demonstrated that the principal component analysis can be easily outperformed by the linear correlation and the all-possible-regressions methods, the latter yielding a root-mean-squared error (RMSE) of 6.77 h when applied with three catchment descriptors. The most interesting finding of the dimension reduction is that Tr is strongly connected to field capacity in the region of study. The performance of the clustering methods varies considerably (RMSE = 5.05–12.03 h). The best overall performance comes from the residual approach (RMSE = 8.14 h on average). It is shown that several of the methods outperform the grouping based on geographical regions, however, the estimation error is reduced only in a few cases when compared to the regional estimation (i.e., one cluster) method. Clusters created based on catchment width yields the best results, resulting in RMSE values of 5.80 and 5.77 h (with two and four clusters, respectively). The comparison of the new and the existing empirical equations clearly demonstrated that the estimation of Tr needs improvement in Hungary, while the application of more than two clusters is unwarranted for the study region.

Design flood estimation for global river networks based on machine learning models

Abstract. Design flood estimation is a fundamental task in hydrology. In this research, we propose a machine-learning-based approach to estimate design floods globally. This approach involves three stages: (i) estimating at-site flood frequency curves for global gauging stations using the Anderson–Darling test and a Bayesian Markov chain Monte Carlo (MCMC) method; (ii) clustering these stations into subgroups using a K-means model based on 12 globally available catchment descriptors; and (iii) developing a regression model in each subgroup for regional design flood estimation using the same descriptors. A total of 11 793 stations globally were selected for model development, and three widely used regression models were compared for design flood estimation. The results showed that (1) the proposed approach achieved the highest accuracy for design flood estimation when using all 12 descriptors for clustering; and the performance of the regression was improved by considering more descriptors during training and validation; (2) a support vector machine regression provided the highest prediction performance amongst all regression models tested, with a root mean square normalised error of 0.708 for 100-year return period flood estimation; (3) 100-year design floods in tropical, arid, temperate, cold and polar climate zones could be reliably estimated (i.e. <±25 % error), with relative mean bias (RBIAS) values of −0.199, −0.233, −0.169, 0.179 and −0.091 respectively; (4) the machine-learning-based approach developed in this paper showed considerable improvement over the index-flood-based method introduced by Smith et al. (2015, https://doi.org/10.1002/2014WR015814) for design flood estimation at global scales; and (5) the average RBIAS in estimation is less than 18 % for 10-, 20-, 50- and 100-year design floods. We conclude that the proposed approach is a valid method to estimate design floods anywhere on the global river network, improving our prediction of the flood hazard, especially in ungauged areas.

Hydrological characteristics of Australia: relationship between surface flow, climate and intrinsic catchment properties

Streamflow and baseflow dynamics are driven by complex, interconnected catchment properties. A national study was conducted to assess the relationship between surface flow, climate and intrinsic catchment attributes in Australia. Subcatchments were delineated based on Horton's 5th stream order and were characterised by identifying parameters that influence streamflow and flood behaviour. Because observational datasets like rainfall and streamflow commonly have a non-normal distribution, the method of L-moments was applied to several time series. Surface hydrology and baseflow patterns were represented by twenty indices, which were statistically summarised via principal component (PC) analysis, yielding six PCs. Forty catchment descriptors from the themes of climate, topography, surface condition and hydrogeology were used to investigate their link with runoff patterns. Among these is the land surface value, a newly defined index incorporating soil properties and land use to estimate the capacity for water infiltration. All metrics were explored via correlation and regression analysis against the surface hydrology PCs and their influence on runoff discussed. The predictive skill of the regression models is improved when non-perennial waterways are excluded. Although rainfall characteristics dominate streamflow behaviour, topographical and surface conditions also greatly impact on runoff, especially during low-flow periods.