Search

Groundwater (GW) is commonly conceptualized as causally linked to streamflow (SF). However, confirming where and how it occurs is challenging given the expense of experimental field monitoring. Therefore, hydrological modeling and water management often rely on expert knowledge to draw causality between SF and GW. This paper investigates the potential of convergent cross‐mapping (CCM) to identify causal interactions between SF and GW head. Widely used in ecology, CCM is a nonparametric method to identify causality in nonlinear dynamic systems. To apply CCM between variables the only required inputs are time‐series data (stream gauge and bore), so it may be an attractive alternative or complement to expensive field‐based studies of causality. Three upland catchments across different hydrogeologic settings and climatic conditions in Victoria, Australia, are adopted as case studies. The outputs of the method seem to largely agree with a priori perceptual understanding of the study areas and offered additional insights about hydrological processes. For instance, it suggested weaker SF‐GW interactions during and after the Millennium Drought than in the previous wet periods. However, we show that CCM limitations around seasonality, data sampling frequency, and long‐term trends could impact the variability and significance of causal links. Hence, care must be taken while physically interpreting the causal links suggested by CCM. Overall, this study shows that CCM can provide valuable causal information from common hydrological time‐series, which is relevant to a wide range of applications, but it should be used and interpreted with care and future research is needed. Plain Language Summary: It is common in hydrology to assume that the water below the ground (in aquifers) interacts with the rivers. However, it is challenging to identify where and how it occurs with the available data and financial resources. Therefore, we often use computer models that represent these interactions according to the knowledge of experts. This paper investigates the potential of convergent cross‐mapping (CCM), a method that is widely used in ecology, to identify these interactions. This method only requires time‐series data (e.g., streamflow and groundwater level data) and represents an attractive alternative or complement to expensive field‐based studies. The results of the analysis appear to agree with an initial understanding of the aquifer‐river interactions in three study areas in Victoria, Australia, and offered additional insights. For instance, it suggests weaker interactions during and after the Millennium Drought than in the previous wet periods. However, variability and significance of the causal links inferred by CCM can be due to some limitations of the method. Hence, care must be taken while physically interpreting the results. Overall, CCM can unlock valuable information from common data, which is relevant to hydrology, but it should be used and interpreted with care and future research is needed. Key Points: Convergent cross‐mapping (CCM) is a nonparametric method to identify causality in nonlinear dynamic systems that is widely used in ecologyCCM results suggest heterogeneous causal links that are largely consistent with perceptual understanding of processesCCM seems widely applicable in hydrology, but also has limitations and thus should be used and interpreted with care [ABSTRACT FROM AUTHOR]

In this Reply, we address the concerns of Knoben and Clark (2023, https://doi.org/10.1029/2022WR034294) or KC23 that "the assumptions needed to effectively use difference‐based variance estimation methods are not always met by hourly streamflow records." There should be little doubt that the assumptions of our difference‐based estimator will sometimes be violated in hourly streamflow records but the results from de Oliveira and Vrugt (2022, https://doi.org/10.1029/2022wr032263) and confirmed by the findings in our Reply show that such violations are sporadic enough not to affect much the error variance estimates. Snowmelt as pointed out by KC23 (https://doi.org/10.1029/2022WR034294) may not have received sufficient attention in our paper, yet their 365‐day record is simply not long enough to demonstrate bias of our discharge error variance estimates (and their dependence on flow level). This would require analysis of a much longer, multi‐year, streamflow record of a snowmelt‐driven watershed. The snowmelt catchment analyzed in dOV22 (https://doi.org/10.1029/2022wr032263) did not demonstrate bias in the discharge error variance estimates. We also provide additional clarification on the interpretation of the variance estimates obtained with the nonparametric estimator, and discuss the main issues in the test case presented in Knoben and Clark (2023, https://doi.org/10.1029/2022WR034294)). Key Points: We address the concerns of Knoben and Clark (2023) on the use of the nonparametric estimator with hourly discharge dataWe provide further clarification on the interpretation of the variance estimates obtained with the nonparametric estimator [ABSTRACT FROM AUTHOR]

Hydrologic signatures are quantitative metrics that describe a streamflow time series. Examples include annual maximum flow, baseflow index and recession shape descriptors. In this paper, we use machine learning (ML) to learn encodings that are optimal ML equivalents of hydrologic signatures, and that are derived directly from the data. We compare the learned signatures to classical signatures, interpret their meaning, and use them to build rainfall‐runoff models in otherwise ungauged watersheds. Our model has an encoder–decoder structure. The encoder is a convolutional neural net mapping historical flow and climate data to a low‐dimensional vector encoding, analogous to hydrological signatures. The decoder structure includes stores and fluxes similar to a classical hydrologic model. For each timestep, the decoder uses current climate data, watershed attributes and the encoding to predict coefficients that distribute precipitation between stores and store outflow coefficients. The model is trained end‐to‐end on the U.S. CAMELS watershed data set to minimize streamflow error. We show that learned signatures can extract new information from streamflow series, because using learned signatures as input to the process‐informed model improves prediction accuracy over benchmark configurations that use classical signatures or no signatures. We interpret learned signatures by correlation with classical signatures, and by using sensitivity analysis to assess their impact on modeled store dynamics. Learned signatures are spatially correlated and relate to streamflow dynamics including seasonality, high and low extremes, baseflow and recessions. We conclude that process‐informed ML models and other applications using hydrologic signatures may benefit from replacing expert‐selected signatures with learned signatures. Key Points: We built an encoder–decoder network to learn an optimal equivalent of hydrologic signaturesUsing learned signatures as input to a process‐informed ML model improves prediction accuracy over benchmark configurationsWe interpret learned signatures by correlation with classical signatures and by sensitivity analysis of their impact on model store dynamics [ABSTRACT FROM AUTHOR]

The need for and the use of different data assimilation techniques to improve the quality of streamflow forecast is now well established. In this paper, the goal is to demonstrate the power of a new class of methods known as the forward sensitivity method (FSM) which is based on the temporal evolution of model sensitivities with respect to the control variables consisting of initial conditions and parameters. FSM operates in two phases: the first phase provides a simple algorithm for placing observations at or near where the square of forward sensitivities attains their maximum values. Using only this selected subset of observations in a weighted least squares method, the second phase then provides an estimate of the unknown elements of the control variables. In this paper, FSM based assimilation is applied to a simple class of two parameter model in a medium‐sized agriculture dominant watershed lying in the Krishna River Basin, India. Four assimilation scenarios were tested to determine the effect of assimilating only sensitive observations as well as the impact of temporally evolving initial condition sensitivity. Sensitivity results showed that observations during the monsoon time alone are enough for assimilation purposes, which has helped in reducing the computational time greatly. Assimilation and forecast results also indicated that the scenarios which assimilated only sensitive observations are better in estimating daily streamflow. From the obtained results, it is concluded that FSM based assimilation has significant potential to improve the streamflow simulations, especially in places where data availability remains a major challenge. Key Points: The potential of assimilating only sensitive soil moisture observations on streamflow forecast was demonstrated using a novel forward sensitivity method (FSM) methodImpact of temporally evolving sensitivity of initial condition on hydrological model performance during assimilation was determinedThe sensitivity‐based soil moisture assimilation was effective up to 45 lead days during streamflow forecast [ABSTRACT FROM AUTHOR]

Hydrologic variability can present severe financial challenges for organizations that rely on water for the provision of services, such as water utilities and hydropower producers. While recent decades have seen rapid growth in decision‐support innovations aimed at helping utilities manage hydrologic uncertainty for multiple objectives, support for managing the related financial risks remains limited. However, the mathematical similarities between multi‐objective reservoir control and financial risk management suggest that the two problems can be approached in a similar manner. This paper demonstrates the utility of Evolutionary Multi‐Objective Direct Policy Search for developing adaptive policies for managing the drought‐related financial risk faced by a hydropower producer. These policies dynamically balance a portfolio, consisting of snowpack‐based financial hedging contracts, cash reserves, and debt, based on evolving system conditions. Performance is quantified based on four conflicting objectives, representing the classic tradeoff between "risk" and "return" in addition to decision‐makers' unique preferences toward different risk management instruments. The dynamic policies identified here significantly outperform static management formulations that are more typically employed for financial risk applications in the water resources literature. Additionally, this paper combines visual analytics and information theoretic sensitivity analysis to improve understanding about how different candidate policies achieve their comparative advantages through differences in how they adapt to real‐time information. The methodology presented in this paper should be applicable to any organization subject to financial risk stemming from hydrology or other environmental variables (e.g., wind speed, insolation), including electric utilities, water utilities, agricultural producers, and renewable energy developers. Key Points: Reservoir control and financial risk management share a common multi‐objective decision structure and can be optimized using similar methodsEvolutionary Multi‐Objective Direct Policy Search is used to develop financial risk management policies for a hydropower producerInformation theoretic sensitivity analysis and visual analytics are used to build intuition about how policies adapt to changing conditions [ABSTRACT FROM AUTHOR]

Occurrences of extreme events, especially floods, have become more frequent and severe in the recent past due to the global impacts of climate change. In this context, possibilities for generating a near‐accurate streamflow forecast at higher lead times, which could be utilized for developing a reliable flood warning system to minimize the effects of extreme events, are highly important. This paper aims to investigate the potential of a novel hybrid modeling framework that couples the random forest algorithm, particle filter, and the HBV model for improving the overall accuracy of forecasts at higher lead times through the dynamic error correction schematic. The new framework simulates an ensemble of streamflow for estimating uncertainty associated with the predictions and is applied across two snow‐fed Himalayan rivers: the Beas River in India and the Sunkoshi River in Nepal. Several statistical indices along with graphical performance indicators were used for assessing the accuracy of the model performance and associated uncertainty. The modeling framework achieved the Nash Sutcliffe Efficiency of 0.94 and 0.98 in calibration and 0.95 and 0.99 in validation for the Beas and Sunkoshi river basin respectively for a 7‐day ahead forecast. Thus, the proposed framework can be considered as a promising tool having reasonably good performance in forecasting streamflow at a higher lead time. Key Points: Hybrid hydrological model integrates process‐based model with machine learning algorithm through data assimilation techniqueDynamic error correction framework capable of improving the streamflow forecast at longer lead time is proposedOverall the developed framework improves the forecast accuracy along with quantifying the model prediction uncertainty [ABSTRACT FROM AUTHOR]

Stochastic methods have been typically used for the design and operations of hydraulic infrastructure. They allow decision makers to evaluate existing or new infrastructure under different possible scenarios, giving them the flexibility and tools needed in decision making. In this paper, we present a novel stochastic streamflow simulation approach able to replicate both temporal and spatial dependencies from the original data in a multi‐site basin context. The proposed model is a multi‐site extension of the modified Fractional Gaussian Noise (mFGN) model which is well‐known to be efficient to maintain periodic correlation for several time lags, but presents shortcomings in preserving the spatial correlation. Our method, called Weighted‐mFGN (WmFGN), incorporates spatial dependency into streamflows simulated with mFGN by relying on the Cholesky decomposition of the spatial correlation matrix of the historical streamflow records. As the order in which the decomposition steps are performed (temporal then spatial, or vice‐versa) affects the performance in terms of preserving the temporal and spatial correlation, our method searches for an optimal convex combination of the resulting correlation matrices. The result is a Pareto‐curve that indicates the optimal weights of the convex combination depending on the importance given by the user to spatial and temporal correlations. The model is applied to a number of river basins in Chile, where the results show that the WmFGN approach maintains the qualities of the single‐site mFGN, while significantly improving spatial correlation. Key Points: Weighted modified Fractional Gaussian Noise model improves spatial correlations without significantly sacrificing temporal correlationThe method searches for an optimal convex combination of the spatial and temporal correlation matrices according on the user's priorityOur results in several Chilean basins demonstrate that Weighted‐modified Fractional Gaussian Noise (WmFGN) represents a significant improvement over the mFGN in preserving correlations [ABSTRACT FROM AUTHOR]

River networks are not steady blue lines drawn in a map, since they continuously change their shape and extent in response to climatic drivers. Therefore, the flowing length of rivers (L) and the corresponding catchment‐scale streamflow (Qsur) co‐evolve dynamically. This paper analyzes the relationship between the wet channel length and the streamflow of a river basin, formulating a general analytical model that includes the case of temporarily dry outlets. In particular, the framework relaxes the common assumption that when the discharge at the outlet tends to zero the upstream flowing length approaches zero. Different analytical expressions for the L(Qsur) law are derived for the cases of (a) a perennial outlet; (b) a non‐perennial outlet that dries out only when the whole network is dry; and (c) a temporarily dry outlet, that experiences surface flow for less time than other network nodes. In all cases, the shape of the L(Qsur) relationship is controlled by the distribution of the specific subsurface discharge capacity along the network. For temporarily dry outlets, however, the relation between L and Qsur might depend on an unknown shifting factor. Three real‐world examples are presented to demonstrate the flexibility and the robustness of the theory. Our results indicate that the whole shape of the L(Qsur) relation might not be empirically observable if a significant fraction of the network is perennial or some reaches in the network experience surface flow for longer than the discharge gauging station. The study provides a basis for integrating empirical L(Qsur) data gathered in diverse sites. Key Points: Active network length (L) versus catchment discharge (Qsur) relations are extended to temporarily dry outletsAll the L(Qsur) relations represent translations/truncations of the spatial cumulative distribution of the specific subsurface capacityThe L(Qsur) law is not fully visible if the outlet is less persistent than other network nodes (as in a spring‐fed stream drying upward) [ABSTRACT FROM AUTHOR]

Optimization of spatially consistent parameter fields is believed to increase the robustness of parameter estimation and its transferability to ungauged basins. The current paper extends previous multi‐objective and transferability studies by exploring the value of both multi‐basin and spatial pattern calibration of distributed hydrologic models as compared to single‐basin and single‐objective model calibrations, with respect to tradeoffs, performance and transferability. The mesoscale Hydrological Model (mHM) is used across six large central European basins. Model simulations are evaluated against streamflow observations at the basin outlets and remotely sensed evapotranspiration patterns. Several model validation experiments are performed through combinations of single‐ (temporal evaluation through discharge) and multi‐objective (temporal and spatial evaluation through discharge and spatial evapotranspiration patterns) calibrations with holdout experiments saving alternating basins for model evaluation. The study shows that there are very minimal tradeoffs between spatial and temporal performance objectives and that a joint calibration of multiple basins using multiple objective functions provides the most robust estimations of parameter fields that perform better when transferred to ungauged basins. The study indicates that particularly the multi‐basin calibration approach is key for robust parametrizations, and that the addition of an objective function tailored for matching spatial patterns of ET fields alters the spatial parameter fields while significantly improving the spatial pattern performance without any tradeoffs with discharge performance. In light of model equifinality, the minimal tradeoff between spatial and temporal performance shows that adding spatial pattern evaluation to the traditional temporal evaluation of hydrological models can assist in identifying optimal parameter sets. Plain Language Summary: Hydrological models typically require local observations of river flow to calibrate the models and test their predictive capability. This limits the possibility for predictions in ungauged basins. This study used holdout tests to investigate the robustness of hydrological predictions for ungauged basins. Particularly we investigate how adding more basins and observed spatial patterns of evapotranspiration in the calibration of these models impact this robustness and transferability of model parameters to basins not used for calibration. Results show that transferability and spatial consistency of parameters increase when adding more basins and spatial pattern observations. Key Points: Clear improvements in simulated spatial patterns can be achieved with very limited tradeoff with discharge performanceMulti‐basin and spatial pattern‐based calibration improve parameter realism and transferability to ungauged basins [ABSTRACT FROM AUTHOR]

The basic structure of the scientific method-at least in its idealized form-is widely championed as a recipe for scientific progress, but the day-to-day practice may be different. Here, we explore the spectrum of current practice in hypothesis formulation and testing in hydrology, based on a random sample of recent research papers. This analysis suggests that in hydrology, as in other fields, hypothesis formulation and testing rarely correspond to the idealized model of the scientific method. Practices such as 'p-hacking' or 'HARKing' (Hypothesizing After the Results are Known) are major obstacles to more rigorous hypothesis testing in hydrology, along with the well-known problem of confirmation bias-the tendency to value and trust confirmations more than refutations-among both researchers and reviewers. Nonetheless, as several examples illustrate, hypothesis tests have played an essential role in spurring major advances in hydrological theory. Hypothesis testing is not the only recipe for scientific progress, however. Exploratory research, driven by innovations in measurement and observation, has also underlain many key advances. Further improvements in observation and measurement will be vital to both exploratory research and hypothesis testing, and thus to advancing the science of hydrology. [ABSTRACT FROM AUTHOR]

In this paper, we introduce a relatively simple data‐driven method for the representation of the uncertainty in daily discharge records. The proposed method relies only on hourly discharge data and takes advantage of a nonparametric difference‐based estimator in the characterization of random errors in discharge time series. We illustrate with corrupted streamflow data that the nonparametric estimator provides an accurate characterization of the nature (homoscedastic or heteroscedastic) and magnitude of these errors. In addition, we demonstrate the practical usefulness of the estimator using discharge time series of 500+ watersheds of the Catchment Attributes and MEteorology for Large‐sample Studies data set. This analysis reveals that the magnitude of errors of aleatory nature in the investigated discharge records is rather small (less than 3% for 80% of the records). We then combine the effect of random errors and measurement frequency into a daily variance estimate, which serves as input to a streamflow generation approach. This procedure produces replicates of the discharge record which portray accurately the assigned streamflow uncertainty, preserve key statistical properties of the discharge record and are hydrologically realistic. The proposed method facilitates Bayesian analysis and supports tasks such as model diagnostics, data assimilation, uncertainty quantification and regionalization. Key Points: We present a model‐agnostic method for the generation of replicates of discharge recordsThe nonparametric estimator provides an accurate estimate of random errors, which for the investigated discharge records are rather smallThe replicates portray accurately the assigned streamflow uncertainty and preserve statistical and hydrologic properties [ABSTRACT FROM AUTHOR]

Reliable urban flood modeling is highly demanded in emergency response, risk management, and urban planning related to urban flooding. In this paper, the Storm Water Management Model (SWMM) is adapted to simulate urban rainfall‐runoff and pipe drainage processes within the Dominant river tracing‐Routing Integrated with VIC Environment (DRIVE) model which accounts for natural river basin runoff generation and routing processes. The integrated DRIVE‐SWMM model (referred to as DRIVE‐Urban) allows to explicitly delineate the mass‐energy interactions between urban drainage system (e.g., pipes and dikes) and river networks. This presents a further step model development for accurate urban flooding prediction which is lacking in existing urban flood models and traditional hydrological models. The validity of the DRIVE‐Urban model is evaluated for three case studies in Haikou City, China, with camera observations of street inundation during typhoon landfalls and heavy rainfall events. The results show that the DRIVE‐Urban model successfully captures 62%, 69%, and 77% of the total observed inundated road‐sections for the three cases respectively. The third case study with severe flooding situation shows that the DRIVE‐Urban performance is further improved when given reliable river and tidal level information, indicating the importance of integrating river‐basin with urban hydrological and hydraulic modeling. Plain Language Summary: This study mainly focuses on the development of an urban flood model (Dominant river tracing‐Routing Integrated with VIC Environment, DRIVE‐Urban), which couples the river‐basin and urban hydrological modeling, with a great potential for large‐scale real‐time urban flood monitoring and forecasting through the current Global Flood Monitoring System. Two inundation algorithms are developed within the DRIVE‐Urban model. One is for the river inundation caused urban flooding, and the other one is for the urban inundation caused by manhole overflow or excessive runoff drainage over the city land surface. Multi‐source observations including the closed‐circuit television video, citizen photos and municipal reports are collected for model validation. The results show that the DRIVE‐Urban model simulates well the different type of urban inundations at the 5‐m high spatial resolution with a promising hit success (i.e., POD >60%) according to three case studies. In particular, the results for the typhoon Kalmaegi case show that the model performance can be improved significantly by coupling the upstream river flow simulation with the urban runoff‐routing process, which validates our strategy of including the upstream river‐basin hydrological process in the urban flood modeling. Key Points: The Dominant river tracing‐Routing Integrated with VIC Environment (DRIVE)‐Urban flood model is developed by coupling river basin and urban hydrological and hydraulic processTwo 2D inundation algorithms are implemented within the model to simulate the manhole inundating and river overflow flooding respectivelyThe validity of the DRIVE‐Urban model is well established through three case studies with observations during typhoon and rainfall events [ABSTRACT FROM AUTHOR]

Multivariate data assimilation (DA), a novel way to couple big data with land surface models, was extensively employed in forecasting‐reanalyzing systems (FRSs), for example, ECMWF and GLDAS. Meanwhile, most (distributed) hydrological models, like soil and water assessment tool (SWAT), have not been equipped with straightforward ways to link to DA algorithms. Therefore, it is one of the main barriers to utilizing such hydrological models in FRSs. This paper deals with multivariate DA into SWAT (DA‐SWAT), which is complicated since the original model does not provide full access to the models' initial conditions (ICs) at the hydrologic response unit (HRU) scale. The preceding DA‐SWAT works commonly used an integrated approach in which the DA and SWAT codes were implemented in the same programming environment. We discuss how this approach complicates and prevents the application of DA‐SWAT in multivariate, multimodel, and multisensor systems. Accordingly, we proposed a new approach for DA‐SWAT by which SWAT can be perfectly linked with any DA algorithm of interest coded in any desired programming environment. Our framework utilizes input/output text files to access ICs and to link DA with SWAT. Moreover, we designed some univariate and multivariate scenarios for assimilating in situ streamflow measurement and MODIS's snow cover fraction (SCF) data, which has not yet been focused on in the SWAT calibration context. Results show that compared to the univariate assimilation of streamflow (SCF), the multivariate assimilation mitigates the equifinality problem and more accurately estimates SCF (streamflow) by improving NS and PBIAS measures with the differences of 0.4 (0.86), 12% (64%), respectively. Key Points: We propose a new approach to link SWAT with data assimilation (DA) algorithms based on source code modificationSnow cover fraction (SCF) data of MODIS are assimilated to estimate HRU‐scaled snow‐related parameters of SWATBivariate streamflow and MODIS SCF data assimilation mitigates the equifinality problem [ABSTRACT FROM AUTHOR]

Streamflow prediction is a long‐standing hydrologic problem. Development of models for streamflow prediction often requires incorporation of catchment physical descriptors to characterize the associated complex hydrological processes. Across different scales of catchments, these physical descriptors also allow models to extrapolate hydrologic information from one catchment to others, a process referred to as "regionalization". Recently, in gauged basin scenarios, deep learning models have been shown to achieve state of the art regionalization performance by building a global hydrologic model. These models predict streamflow given catchment physical descriptors and weather forcing data. However, these physical descriptors are by their nature uncertain, sometimes incomplete, or even unavailable in certain cases, which limits the applicability of this approach. In this paper, we show that by assigning a vector of random values as a surrogate for catchment physical descriptors, we can achieve robust regionalization performance under a gauged prediction scenario. Our results show that the deep learning model using our proposed random vector approach achieves a predictive performance comparable to that of the model using actual physical descriptors. The random vector approach yields robust performance under different data sparsity scenarios and deep learning model selections. Furthermore, based on the use of random vectors, high‐dimensional characterization improves regionalization performance in gauged basin scenario when physical descriptors are uncertain, or insufficient. Key Points: We propose a random vector strategy to serve as a surrogate to physical descriptors for regionalization under gauged scenariosThe deep learning model using random vectors yields comparable performance to the model using physical descriptors in gauged scenariosHigh‐dimensional characterization helps distinguish catchments and thus helps catchments learn from each other [ABSTRACT FROM AUTHOR]

Despite long‐standing efforts, hydrologists still lack robust tools for calibrating land surface model (LSM) streamflow estimates within ungauged basins. Using surface soil moisture estimates from the Soil Moisture Active Passive Level 4 Soil Moisture (L4_SM) product, precipitation observations, and streamflow gauge measurements for 617 medium‐scale (200–10,000 km2) basins in the contiguous United States, we measure the temporal (Spearman) rank correlation between antecedent (i.e., pre‐storm) surface soil moisture (ASM) and the storm‐scale runoff coefficient (RC; the fraction of storm‐scale precipitation accumulation converted into streamflow). In humid and semi‐humid basins, this rank correlation is shown to be sufficiently strong to allow for the substitution of storm‐scale RC observations (available only in basins that are both lightly regulated and gauged) with high‐quality ASM values (available quasi‐globally from L4_SM) in streamflow calibration procedures. Using this principle, we define a new, basin‐wise LSM streamflow calibration approach based on L4_SM alone and successfully apply it to identify LSM configurations that produce a high rank correlation with observed RC. However, since the approach cannot detect RC bias, it is less successful in identifying LSM configurations with low mean‐absolute error. Plain Language Summary: Accurately forecasting the fraction of rainfall that runs off into streams, as opposed to infiltrates into the soil, is critical for flash‐flood prediction, water‐resource monitoring, and tracking the transport of nutrients from agricultural fields into local waterways. Such forecasting is typically performed by hydrologic models that attempt to represent the physical processes responsible for surface runoff generation. However, to provide accurate streamflow forecasts, these models typically need to be calibrated against actual streamflow observations. This is problematic given the relatively poor, and declining, global availability of stream gauges. This paper presents a novel model calibration strategy that uses soil moisture from remote sensing and numerical modeling in place of streamflow observations during calibration. This transition has significant practical advantages because, unlike streamflow observations, the soil moisture data are continuously available across space. Our results demonstrate that this new approach can significantly improve hydrologic models within humid and semi‐humid basins lacking sufficient ground‐based instrumentation for traditional streamflow calibration. Key Points: Hydrologic models still perform poorly in ungauged basinsIn relative humid areas, the correlation between pre‐storm soil moisture and the storm‐scale runoff coefficient approaches unitySuch strong correlation can be exploited to improve the precision of runoff coefficient estimates provided by models in ungaged basins [ABSTRACT FROM AUTHOR]

Global flood models integrate flood maps of constant probability in space, ignoring the correlation between sites and thus potentially misestimating the risk posed by extreme events. Stochastic flood models alleviate this issue through the simulation of flood events with a realistic spatial structure, yet their proliferation at large scales has historically been inhibited by data quality and computer availability. In this paper, we show, for the first time, the efficacy of modeled river discharge reanalyses in the characterization of flood spatial dependence in the absence of a dense stream gauge network. While global hydrological models may show poor correspondence with absolute observed river flows, we find that the rate at which they can simulate the joint occurrence of relative flow exceedances at two given locations is broadly similar to when a gauge‐based statistical model is used. Evidenced over the United States, flood events simulated using observed gauge data from the U.S. Geological Survey versus those generated using modeled streamflows have similar (i) distributions of site‐to‐site correlation strength, (ii) relationships between event size and return period, and, importantly, (iii) loss distributions when incorporated into a continental‐scale flood risk model. Extremal dependence is generally quantified less accurately on larger rivers, in arid climates, in mountainous terrain, and for the rarest high‐magnitude events. However, local‐scale errors are shown to broadly cancel each other out when combined, producing an unbiased flood spatial dependence model. These findings suggest that building accurate stochastic flood models worldwide may no longer be a distant aspiration. Plain Language Summary: Global flood risk is commonly estimated through flood inundation maps with a defined probability of occurrence. These flood simulations have a key drawback in that they fail to capture the spatial patterns exhibited during real flood events, instead modeling the same probability of flooding on every river at once. Solutions which rely on networks of gauged river flow observations will necessarily break down in the majority of the world's regions which lack such a resource. In this paper, we use historic river flows simulated by global rainfall‐runoff models (rather than observed flows) into a statistical model which captures the spatial correlation of flow extremes. If we examine the relative flow exceedance probabilities from these hydrological models rather than the volumetric flow values, flood events are generated which exhibit similar characteristics to those when gauged flow observations are used. Crucially, the simulation‐ and observation‐generated flood events produce near‐identical losses to buildings in the United States. The implications of this are that true stochastic flood risk models, which account for spatial dependence, can proliferate globally via the generation of realistic flood event sets from hydrological models. Key Points: Large‐scale flood hazard models typically neglect to represent the spatial dependence of real flood eventsRelative flow exceedances simulated by the fusion of global hydrological with statistical models reproduce gauge‐driven flood event setsAt the continental scale, key characteristics of a flood risk model are indistinguishable when driven with observed versus modeled flow data [ABSTRACT FROM AUTHOR]

Streamflow elasticity to precipitation, defined as the percent change of streamflow resulting from a 1% change in precipitation, is sometimes used as an alternative to rainfall‐runoff models in climate impact analyses. Elasticity is usually estimated from long streamflow and precipitation series aggregated at annual time steps while the climate impact analyses are usually geared toward changes at decadal scales. The purpose of this paper is therefore to understand how the elasticity depends on the aggregation time scale and the process controls of such a dependence. We analyze streamflow records of 7,053 catchments around the world over the period 1950–2016, and select 5,327 with reliable elasticity estimates for aggregation time ranging from 13 to 60 months. We find a significant scaling of streamflow elasticity to precipitation with aggregation time in 66% of the catchments which is much larger than expected by chance. Positive scaling occurs much more frequently than negative scaling. More arid/less rainy catchments, less forested catchments and catchments with a large base flow contribution to streamflow are more frequently characterized by a positive scaling. A random forest classification model identifies aridity index, latitude, mean annual precipitation, the potential evapotranspiration seasonality, the base flow index and the precipitation seasonality as relevant explanatory variables of the scaling. We interpret the sign of the scaling by non‐linear runoff generation in arid regions, by the effect of climate modes and snow processes, and by the regulation capacity of vegetation to transpire more water if the past years were wet. It is suggested to use decadal elasticities instead of annual elasticities in climate impact analyses in order to account for their scaling behavior. Key Points: Streamflow elasticity to precipitation depends significantly on aggregation time in 66% out of the 5,327 global catchmentsCatchments that are arid or little forested, or possess gentle catchment slope tend to exhibit positive scalingDecadal streamflow elasticity to precipitation should be used instead of annual elasticities in climate impact analyses [ABSTRACT FROM AUTHOR]

Concentration‐discharge (C‐Q) relationships are frequently used to understand the controls on material export from watersheds. These analyses often use a log‐log power‐law function (C = aQb) to determine the relationship between C and Q. Use of the power‐law in C‐Q analyses dates to two seminal papers by Francis Hall (1970, https://doi.org/10.1029/WR006i003p00845) and Francis Hall (1971, https://doi.org/10.1029/WR007i003p00591), where he compared six increasingly complex hydrological models, concluding the power‐law had the greatest explanatory power. Hall's analyses and conclusions, however, were based on a limited data set, with assumptions regarding water volume and storage, and used simple model selection criteria. While the power‐law is applied widely, it has not been rigorously tested and evaluated in over 50 years. We reexamined Hall's original models across time scales using 8 years of high‐frequency and weekly specific conductance data and evaluated model performance using more sophisticated model selection criteria. While we found the power‐law analysis remains one of the best performing models, other models performed equally as well including the log‐linear functional form. Model performance was similar at the sub‐daily to weekly scale but varied with sampling method. More complex models performed poorly relative to simpler models and tended to underpredict concentration at flow extremes due to constraints in fitting model parameters to the observed data. While we conclude, based on the data analyzed here, that the power‐law remains a suitable model for C‐Q analyses, opportunities exist to refine and differentiate among C‐Q models based on underlying assumptions of data distribution, recession analyses, and for applying models to reactive solutes. Plain Language Summary: Understanding how solute concentrations respond to changes in river flow remains a fundamental challenge in water resources science. Evaluating the relationship between solute concentrations and river flow (or discharge) can provide insight into how watersheds are structured and how they function. Concentration‐discharge (C‐Q) relationships can be used to estimate the rate of material export from terrestrial landscapes to surface waters. Many C‐Q analyses use a power‐law analysis and model to determine the response of C to variation in Q; yet this model and its embedded assumptions have not been rigorously tested following the development of water quality sensors that provide high‐frequency data. Here we revisit eight mathematical models originally developed by Hall (1970, https://doi.org/10.1029/WR006i003p00845) and Hall (1971, https://doi.org/10.1029/WR007i003p00591) that were initially evaluated with only 36 data points. We reexamine Hall's models using eight years of 15‐min specific conductivity data and find that while the power‐law model still is one of the best models to use in the evaluation of C‐Q relationships. And overall, simpler models outperform more complex models. We discuss many of the assumptions, such as constant load, that underpin C‐Q analyses to demonstrate that future studies could further parameterize C‐Q analyses for more insight on the mechanisms driving solute‐discharge relationships. Key Points: We re‐evaluate equations proposed by Francis Hall to assess concentration‐discharge (C‐Q) relationships using newly available long‐term and high‐frequency data setsAcross time steps we find that log‐log and log‐linear models perform equally well to describe C‐Q relationshipsParametrization of storage‐discharge relationships via recession analyses provides additional insight to C‐Q relationships [ABSTRACT FROM AUTHOR]

It remains an open question whether the state of knowledge on bend hydrodynamics applies to sharp bends in the largest river systems on Earth. This paper reports a field investigation into the hydrodynamics in two consecutive sharp bends of the Middle Yangtze River (MYR), and demonstrates that the reduced‐order model of Blanckaert and de Vriend is able to resolve the secondary flow strength and continuous flow redistribution in rivers of this size to high accuracy. Consequently, the model has been validated over the entire range of scales. These two bends show similarities to those of smaller size in terms of bathymetry, flow redistribution, and secondary flow evolution. The transverse location of the core of maximum velocity along the bends lags behind the thalweg by approximately half a width. A center‐region cell of secondary flow develops over the outer bank near the apex, and gradually decays downstream until it completely vanishes at the bend exit. Both bends are mildly curved from the perspective of hydrodynamic modeling, with a weak interaction between the secondary flow and primary flow, although they are sharply curved from the perspective of geomorphology (width/radius > 0.5). A term‐by‐term analysis of the model indicates that the flow redistribution is primarily controlled by planimetric changes in curvature and by topographic steering, and is slightly influenced by the secondary flow. However, the secondary flow may indirectly affect the flow redistribution by conditioning the bed topography. Important knowledge gaps remain with respect to the coupling between the hydrodynamic and morphological processes. Key Points: Field measurements of 3D flow field in very large sharp bendsPatterns of mean and secondary flows are similar as in smaller systemsThe reduced‐order model can resolve the flow field over the entire range of scales [ABSTRACT FROM AUTHOR]

Despite the recent developments in continental‐scale streamflow and flood inundation modeling frameworks, effects of time‐specific and spatially explicit storage‐release dynamics of numerous dams and reservoirs remain underexplored. This paper fills this knowledge gap by directly inserting operational daily flow release data at 175 dam locations into a streamflow simulation of ∼1.2 million river reaches in the Mississippi River Basin (MRB), and therefore quantifying the effect of these regulations on streamflow and flood inundation extents. Using a streamflow routing model called the Routing Application for Parallel computatIon of Discharge (RAPID) and flood inundation mapping model called AutoRoute, two simulation scenarios were constructed respectively including and excluding the daily flow releases from those dams and reservoirs for a 10‐year period (2005–2014). Flood inundation maps were simulated for peak flow conditions at a ∼10‐m hyper spatial resolution. Kling‐Gupta efficiency (KGE) values show that streamflow model performance considerably improved when reservoirs were included in the modeled system, varying from 2% in the eastern region to 380% in the drier western region. Despite small variation of streamflow model improvement with reservoir release inclusion in the eastern region of the basin, the flow model was able to better capture observed peak flows. For a 1% change in streamflow, we observe a 0.8% change in estimated flood inundation. Comparisons to three observed flood events in the MRB demonstrate that the flood inundation estimates improve when percent change in streamflow is relatively high. Overall, inclusion of reservoir release resulted in substantial improvement in continental‐scale streamflow and flood inundation mapping. Plain Language Summary: Dams and reservoirs are important structures that alter the flow of rivers to provide important services such as flood reduction, hydropower generation, and water storage for irrigation and recreation. We can use computer modeling to simulate the flow of rivers and the operation of dams. Most of the time, researchers approximate the operation of dams using various assumptions. However, we do not fully understand how much the real‐life operation of reservoirs impacts river flow across large spatial extents and by products, such as flood extent. In this study, we combine observed dam releases and computer modeling to replicate river flow in a large river system with many dams. We show how the influence of dams on water movement impacts our modeling of flood maps. Our results reveal that despite the common judgment, small reservoirs can have large influence on water movement. We also show that the dams, to varied degrees, have considerable impact on flood maps when the dams significantly affect the flow of rivers. Key Points: Inclusion of numerous reservoir outflows improves continental‐scale streamflow and hyper‐resolution flood inundation mappingRelative changes to peak flow simulation are substantial downstream of many smaller damsResults show how reservoir effects on streamflow propagate from upstream to downstream [ABSTRACT FROM AUTHOR]

Subseasonal streamflow forecasts, with lead times of 1–30 days, provide valuable information for operational water resource management. This paper introduces the multi‐temporal hydrological residual error (MuTHRE) model to address the challenge of obtaining "seamless" subseasonal forecasts — that is, daily forecasts with consistent high‐quality performance over multiple lead times (1–30 days) and aggregation scales (daily to monthly). The key advance of the MuTHRE model is combining the representation of three temporal characteristics of hydrological residual errors: seasonality, dynamic biases, and non‐Gaussian errors. The MuTHRE model is applied in 11 Australian catchments using the hydrological model GR4J and post processed rainfall forecasts from the numerical weather prediction model ACCESS‐S, and is evaluated against a baseline model that does not model these error characteristics. The MuTHRE model provides "high" improvements (practically significant in the majority of performance stratifications) in terms of reliability: (i) at short lead times (up to 10 days), due to representing non‐Gaussian errors, (ii) stratified by month, due to representing seasonality in hydrological errors, and (iii) in dry years, due to representing dynamic biases in hydrological errors. Forecast performance also improves in terms of sharpness, volumetric bias, and CRPS skill score; these improvements are statistically but not practically significant in the majority of stratifications. Importantly, improvements are consistent across multiple time scales (daily and monthly). This study highlights the benefits of modeling multiple temporal characteristics of hydrological errors and demonstrates the power of the MuTHRE model for producing seamless subseasonal streamflow forecasts that can be utilized for a wide range of applications. Key Points: A multi‐temporal hydrological residual error model is introduced that incorporates seasonality, dynamic biases, and non‐Gaussian errorsSubseasonal streamflow forecasts show "seamless" performance at multiple lead times (1–30 days) and aggregation scales (daily‐monthly)Stratified performance shows large improvements in reliability for shorter lead times, for many months and dry years [ABSTRACT FROM AUTHOR]

Strong earthquakes can cause different kinds of hydrological responses, and several mechanisms have been suggested to explain them. Verification of these mechanisms, however, is often lacking. Here we test some hypotheses with a laboratory experiment, in particular the hypothesis that dynamic strain mobilizes trapped water in the unsaturated zone. We construct a sand chamber, partially saturated with water, and subject it to "seismic" shaking of controlled energy. Pore pressure in the saturated and unsaturated zones is monitored before and after shaking. We identify three distinct mechanisms: consolidation of sediments in the saturated zone, release of capillary water from the capillary fringe, and mobilization of isolated pore water in the unsaturated zone. Each mechanism may cause pore pressure in the saturated zone to suddenly increase with shaking, and each may offer new insights to understand the source of the extra water and other shallow hydrological responses that appear after earthquakes. Plain Language Summary: Earthquakes interact with water in a variety of interesting ways. For instance, during earthquakes, streamflow and groundwater level may increase, as occurred following the recent earthquake swarm in Puerto Rico. The mechanisms for such changes, however, are still unclear. In our paper, we measure the response of groundwater pressure to seismic shaking to test the hypothesis that earthquakes can cause shallow groundwater to flow toward streams. To accomplish this, we built a tall sand column, filled it with water, and instrumented it with sensors that measure how the pressure of water between sand grains (pore pressure) responds when the chamber is shaken by an impact. We found that "trapped" water in the shallow subsurface, which is held in place by adhesive forces between sand and water, can be released to the water table by seismic shaking, where it can then flow toward a stream. Key Points: We test mechanisms for groundwater response to earthquakes by performing laboratory experimentsWe observe consolidation of sediments, mobilization of pore water, and release of capillary water from the capillary fringeThese mechanisms can help explain coseismic increases in groundwater level and streamflow when there are no other apparent water sources [ABSTRACT FROM AUTHOR]

Regions of complex topography and remote wilderness terrain have spatially varying patterns of temperature and streamflow, but due to inherent difficulties of access, are often very poorly sampled. Here we present a data set of distributed stream stage, streamflow, stream temperature, barometric pressure, and air temperature from the Tuolumne River Watershed in Yosemite National Park, Sierra Nevada, California, USA, for water years 2002-2015, as well as a quality-controlled hourly meteorological forcing time series for use in hydrologic modeling. We also provide snow data and daily inflow to the Hetch Hetchy Reservoir for 1970-2015. This paper describes data collected using low-visibility and low-impact installations for wilderness locations and can be used alone or as a critical supplement to ancillary data sets collected by cooperating agencies, referenced herein. This data set provides a unique opportunity to understand spatial patterns and scaling of hydroclimatic processes in complex terrain and can be used to evaluate downscaling techniques or distributed modeling. The paper also provides an example methodology and lessons learned in conducting hydroclimatic monitoring in remote wilderness. [ABSTRACT FROM AUTHOR]

The temporal variability of precipitation and potential evapotranspiration affects streamflow from daily to long‐term scales, but the relative roles of different climate variabilities on streamflow at daily, monthly, annual, and mean annual scales have not been systematically investigated in the literature. This paper developed a new daily water balance model, which provides a unified framework for water balance across timescales. The daily water balance model is driven by four climate forcing scenarios (observed daily climate and observed daily climate with its intra‐monthly, intra‐annual, and inter‐annual variability removed) and applied to 78 catchments. Daily streamflow from the water balance model is aggregated to coarser timescales. The relative roles of intra‐monthly, intra‐annual, and inter‐annual climate variability are evaluated by comparing the modeled streamflow forced with the climate forcings at two consecutive timescales. It is found that daily, monthly, and annual streamflow is primarily controlled by the climate variability at the same timescale. Intra‐monthly climate variability plays a small role in monthly and annual streamflow variability. Intra‐annual climate variability has significant effects on streamflow at all the timescales, and the relative roles of inter‐annual climate variability are also significant to the monthly and mean annual streamflow, which is often disregarded. The quantitative evaluation of the roles of climate variability reveals how climate controls streamflow across timescales. Key Points: Streamflow variations at the daily, monthly, and annual timescales are primarily affected by climate variability at the same timescaleFor mean annual streamflow, intra‐annual variability is the most important climate variability, followed by inter‐annual variabilityIntra‐annual climate variability has significant effects on streamflow at all the timescales [ABSTRACT FROM AUTHOR]

Evidence suggests that catchment state variables such as groundwater can exhibit multiyear trends. This means that their state may reflect not only recent climatic conditions but also climatic conditions in past years or even decades. Here we demonstrate that five commonly used conceptual "bucket" rainfall‐runoff models are unable to replicate multiyear trends exhibited by natural systems during the "Millennium Drought" in south‐east Australia. This causes an inability to extrapolate to different climatic conditions, leading to poor performance in split sample tests. Simulations are examined from five models applied in 38 catchments, then compared with groundwater data from 19 bores and Gravity Recovery and Climate Experiment data for two geographic regions. Whereas the groundwater and Gravity Recovery and Climate Experiment data decrease from high to low values gradually over the duration of the 13‐year drought, the model storages go from high to low values in a typical seasonal cycle. This is particularly the case in the drier, flatter catchments. Once the drought begins, there is little room for decline in the simulated storage, because the model "buckets" are already "emptying" on a seasonal basis. Since the effects of sustained dry conditions cannot accumulate within these models, we argue that they should not be used for runoff projections in a drying climate. Further research is required to (a) improve conceptual rainfall‐runoff models, (b) better understand circumstances in which multiyear trends in state variables occur, and (c) investigate links between these multiyear trends and changes in rainfall‐runoff relationships in the context of a changing climate. Key Points: Environmental state variables such as groundwater display multiyear trends in response to sustained climate anomaliesWe show that five commonly used conceptual "bucket" rainfall runoff models are unable to replicate such trendsBecause of this, the models fail to extrapolate realistically to different climatic conditions, compromising runoff projections Plain Language Summary: It is common in science to use a mental picture or metaphor that simplifies a complex phenomenon. A common metaphor used for a water supply catchment is that of a leaky bucket. When it rains, the bucket fills up; when it does not rain for a while, the bucket empties due to evaporation and water used by trees; and leaking water is like river flow. Computer models based on variants of this metaphor are common and can provide predictions of how much streamflow might occur under future scenarios. This paper explores limitations of the bucket metaphor and associated models. Recently, during a 13‐year drought in Australia, river catchments gradually started to dry up. With each passing year, the depth to groundwater increased gradually as the water used by trees was not replenished by rainfall. We compare this long, slow behavior to that of five commonly used "bucket" models. The models do not show the long, slow drying up—they only show the seasonal ups and downs, and their predictions of streamflow over the drought are poor. This is surprising, and it means we should choose our models carefully and seek out models that can simulate this behavior and its impact on streamflow. [ABSTRACT FROM AUTHOR]

River floods cause extensive losses to economy, ecology, and society throughout the world. They are driven by the space‐time structure of catchment rainfall, which is determined by large‐scale, or even global‐scale, atmospheric processes. The identification of coherent, large‐scale atmospheric circulation structures that determine the moisture transport and convergence associated with rainfall‐induced flooding can help improve its predictability and phenomenology. In this paper, we extend a methodology, used for the analysis of extreme rainfall events, to high streamflow events (HSEs). The approach combines multiple machine learning methods to link HSEs to atmospheric circulation patterns. An application to the German streamflow network using reanalysis data for the period 1960 to 2012 is presented. Daily streamflow from 166 gauges, homogeneously distributed across Germany, are used. Geopotential height fields and integrated vapor transport (IVT) are derived from reanalysis data. An unsupervised neural network, Self Organizing Maps, is applied to geopotential height to identify a finite number of circulation patterns (CPs). Event synchronization between CPs and HSEs is used to establish if they are linked or not. If they are linked, the Event Synchronization method computes the delay between the occurrence of a CP and a HSE. Finally, local logistic regression is used to estimate the probability of occurrence of a HSE, as function of CP and IVT. We demonstrate that our approach is very effective to evaluate HSE probability occurrence across Germany. Key Points: Atmospheric circulation patterns that generate high streamflow events in Germany are identifiedEvent synchronization is used to determine synchronization and delay between atmospheric circulation patterns and high streamflow eventsThe occurrence probability of high streamflow events is conditioned on atmospheric circulation patterns and integrated water vapor transport [ABSTRACT FROM AUTHOR]

Probabilistic modeling of streamflow in ephemeral catchments, where streamflow is frequently zero or negligible, is a major scientific and operational challenge. This paper evaluates the benefits of an explicit treatment of zero flows in the residual error models used for hydrological model calibration and prediction. In this approach, the lower bound of zero for streamflow is implemented using a censoring approach. The explicit approach is compared to a simpler pragmatic approach, which imposes the zero streamflow bound in prediction but not in calibration. Following a theoretical exposition, empirical comparisons are reported using a daily rainfall‐runoff model (GR4J), four residual error schemes (based on log, log‐sinh, and Box‐Cox [BC] transformations with λ = 0.2 and 0.5), 74 Australian catchments with diverse hydroclimatology, and five performance metrics (reliability, precision, bias, proportion of zero flow days, and Continuous Ranked Probability skill score). The key findings are as follows: (1) in mid‐ephemeral catchments (5–50% zero flows) the explicit approach improves predictive performance, especially reliability, through better characterization of residual errors; (2) BC0.2 and BC0.5 schemes are Pareto optimal in mid‐ephemeral catchments (when the explicit approach is used): BC0.2 achieves better reliability and is recommended for probabilistic prediction, whereas BC0.5 attains lower volumetric bias; (3) in low‐ephemeral catchments (<5% zero flows) the pragmatic approach is sufficient; (4) in high‐ephemeral catchments (>50% zero flows) theoretical limitations result in poor performance of these particular explicit and pragmatic approaches, and further development is needed. The findings provide guidance on improving probabilistic streamflow predictions in ephemeral catchments. Key Points: Explicit treatment of zero flows in calibration improves reliability in mid‐ephemeral catchments with 5–50% zero flowsOf the Pareto optimal residual error schemes identified, Box‐Cox with λ = 0.2 favored for probabilistic prediction due to better reliabilityPredictive performance remains poor in catchments with >50% zero flows, due to theoretical limitations of the residual error model [ABSTRACT FROM AUTHOR]

Peatlands cover many low‐lying areas in the tropics. Tropical peatlands are intriguing systems because of their tight coupling between hydrology and carbon storage: They accumulate carbon over thousands of years because of waterlogging, and they remain waterlogged after growing into domed shapes because peat restricts drainage. This feedback between waterlogging and landscape morphology generates landforms with special hydrologic properties that enable simplifications of standard watershed models. In natural tropical peatlands, the water table is always near the surface and infiltration is almost immediate. In addition, water table fluctuations relative to the peat surface are remarkably uniform across tropical peatlands because these peatlands acquire shapes with a uniform topographic wetness index. In this paper, we show that because of these distinctive properties, simple hydrologic models that represent the hydraulic state of a catchment by a scalar quantity that describes total water storage are useful and physically meaningful in tropical peatlands. We demonstrate how to efficiently derive hillslope‐scale parameterizations of transmissivity and specific yield as functions of water table height for a tropical peatland from water table, rainfall, and topographic data. Our findings suggest that natural tropical peatland subcatchments could be usefully modeled as single hydrologic response units for river flow and flood forecasting. Key Points: Hydrology‐ecosystem feedback in tropical peatlands creates large areas with a uniform topographic wetness indexWe show simple ways to estimate transmissivity and specific yield as functions of water level at the hillslope scale in tropical peatlandsWe show that rainfall‐runoff models based on a scalar hydraulic state are especially well suited to natural tropical peatlands [ABSTRACT FROM AUTHOR]

In many hydrological systems, groundwater is pumped from the aquifer onto the land surface, and a fraction of this water subsequently infiltrates to recharge the groundwater system (recirculated groundwater). Tracers that undergo different degrees of reequilibration with the atmosphere during this recirculation process can enable ambient groundwater and recirculated groundwater to be differentiated. In this paper, the recirculated groundwater has been pumped to dewater open pit mines and discharged into ephemeral creeks. Some of this water subsequently recharged back into the aquifer. Chlorofluorocarbon‐12 (CFC‐12), 14C, and 3H are used in a four end‐member mixing analysis to differentiate between (1) ambient groundwater, (2) recirculated groundwater, (3) river recharge from natural flows prior to commencement of mining operations (in 2007), and (4) natural river recharge post‐2007. Sampling of the surface water when discharge of mine water was the only source of river flow enabled the extent of reequilibration of both CFC‐12 and 14C to be accurately determined. Since CFC‐12 reequilibrates more rapidly than 14C, recirculating groundwater had a CFC‐12 concentration that was close to modern, but a 14C activity that was higher than the original groundwater, but less than modern recharge. As 3H does not reequilibrate, it enabled easy differentiation between recirculated groundwater and infiltration of natural creek flows. Uncertainty of end‐member compositions is due to changes in the end‐member concentrations over time in the case of natural river flows, uncertainty in the extent of tracer reequilibration for the groundwater recirculation end‐member, and spatial variations in the composition of the ambient groundwater end‐member. Key Points: Recirculation of water between the land surface and the aquifer can be difficult to identify using ionic tracersThe use of dissolved gases allows quantification of groundwater recirculation due to gas reequilibration on exposure to the atmosphereInfiltration of mine water discharged to creeks is differentiated from natural creek flow using CFC‐12, 14C, and 3H [ABSTRACT FROM AUTHOR]

Simplified flood propagation models are often employed in practical applications for hydraulic and hydrologic analyses. In this paper, we present a new numerical method for the solution of the Linear Parabolic Approximation (LPA) of the De Saint Venant equations (DSVEs), accounting for the space variation of model parameters and the imposition of appropriate downstream boundary conditions. The new model is based on the analytical solution of a cascade of linear diffusive channels in the Laplace Transform domain. The time domain solutions are obtained using a Fourier series approximation of the Laplace Inversion formula. The new Inverse Laplace Transform Diffusive Flood Routing model (ILTDFR) can be used as a building block for the construction of real-time flood forecasting models or in optimization models, because it is unconditionally stable and allows fast and fairly precise computation. [ABSTRACT FROM AUTHOR]

This paper focuses on evaluating the uncertainty of three common regionalization methods for predicting continuous streamflow in ungauged basins. A set of 268 basins covering 1.6 million km2 in the province of Quebec was used to test the regionalization strategies. The multiple linear regression, spatial proximity, and physical similarity approaches were evaluated on the catchments using a leave-one-out cross-validation scheme. The lumped conceptual HSAMI hydrological model was used throughout the study. A bootstrapping method was chosen to further estimate uncertainty due to parameter set selection for each of the parameter set/regionalization method pairs. Results show that parameter set selection can play an important role in regionalization method performance depending on the regionalization methods (and their variants) used and that equifinality does not contribute significantly to the overall uncertainty witnessed throughout the regionalization methods applications. Regression methods fail to consistently assign behavioral parameter sets to the pseudoungauged basins (i.e., the ones left out). Spatial proximity and physical similarity score better, the latter being the best. It is also shown that combining either physical similarity or spatial proximity with the multiple linear regression method can lead to an even more successful prediction rate. However, even the best methods were shown to be unreliable to an extent, as successful prediction rates never surpass 75%. Finally, this paper shows that the selection of catchment descriptors is crucial to the regionalization strategies' performance and that for the HSAMI model, the optimal number of donor catchments for transferred parameter sets lies between four and seven. [ABSTRACT FROM AUTHOR]

Rainfall thresholds for streamflow generation are commonly mentioned in the literature, but studies rarely include methods for quantifying and comparing thresholds. This paper quantifies thresholds in ephemeral streams and evaluates how they are affected by rainfall and watershed properties. The study sites are in southern Arizona, USA; one is hyperarid and the other is semiarid. At both sites rainfall and streamflow were monitored in watersheds ranging from 10−3 to 102 km2. Streams flowed an average of 0–5 times per year in hyperarid watersheds and 3–11 times per year in semiarid watersheds. Although hyperarid sites had fewer flow events, their flow frequency (fraction of rain events causing flow) was higher than in semiarid sites for small (<1 km2) watersheds. At both locations flow frequency decreased with drainage area, but the decrease was steeper in hyperarid watersheds. Watershed mean 60‐min intensity thresholds ranged from 3–13 mm/hr in hyperarid watersheds and 7–16 mm/hr in semiarid watersheds. Higher runoff thresholds and lower flow frequencies in small semiarid watersheds likely relate to greater ground cover and soil development compared to the desert pavement and bedrock surfaces in hyperarid sites. The choice of rain data strongly influenced threshold values; single rain gauges were only adequate for threshold prediction in <1‐km2 watersheds, and incomplete rainfall data led to increases in thresholds with drainage area. We recommend using mean rainfall intensity over the drainage area for threshold analysis because this reduces apparent scale dependence in thresholds caused by incomplete rainfall information. Plain Language Summary: Ephemeral streams in deserts are usually dry, flowing only after heavy rains. Our goal was to determine how much rain is needed for these streams to flow. We studied streams in dry southwestern and in wetter southeastern Arizona, USA. The dry site has mostly bare rock across the landscape, whereas the wetter site has more vegetation, including shrubs, grasses, and oak. In the dry area, small streams flowed 3–5 times per year, and larger streams flowed 0–2 times per year. Streams required 3–13 mm of rain over 1 hr to trigger flow. The small streams flowed more frequently because rain falling on bare rock could rapidly reach the stream. Water was lost into the channel bed before reaching larger streams. In the wetter area, both small and large streams flowed 3–11 times per year. These streams required 7–16 mm of rain over 1 hr to trigger flow. This range of rainfall is slightly higher than for streams in the dry study area, likely because vegetation cover and soil development allow more rainfall to infiltrate into soil before reaching streams. Information on the amount of rainfall needed to trigger streamflow can help with issuing flash flood warnings for ephemeral streams. Key Points: Rainfall thresholds predict streamflow responses with high accuracy in small hyperarid and semiarid watershedsUsing insufficient rain data usually increases threshold values for larger watersheds, leading to apparent scale dependence in thresholdsDeclines in flow frequency and increases in thresholds with drainage area are steeper in hyperarid than in semiarid watersheds [ABSTRACT FROM AUTHOR]

Although not matching the formal definition of the predictive probability distribution, meteorological and hydrological ensembles have been frequently interpreted and directly used to assess flood‐forecasting predictive uncertainty. With the objective of correctly assessing the predictive probability of floods, this paper introduces ways of taking into account the measures of uncertainty provided in the form of ensemble forecasts by modifying a number of well‐established uncertainty postprocessors, such as Bayesian Model Averaging and Model Conditional Processor. The uncertainty postprocessors were developed on the assumption that the future unknown quantity (predictand) is uncertain while model forecasts (predictors) are given, which imply that they are perfectly known. With this in mind, we propose to relax this assumption by considering ensemble predictions, in analogy to measurement errors, as expressions of errors in model predictions to be integrated in the postprocessors coefficients estimation process. The analyses of the methodologies proposed in this work are conducted on a real case study based on meteorological ensemble predictions for the Po River at Pontelagoscuro in Italy. After showing how improper can be the direct use of ensemble predictions to describe the predictive probability distribution, results from the modified postprocessors are compared and discussed. Key Points: Full predictive probability distributions are essential for taking informed decisions thus improving decision makingHydrometeorological models provide ensemble predictions which do not meet the definition of the predictive probability distributionSeveral uncertainty postprocessors, incorporating ensemble spread information, are here introduced to properly incorporating ensemble predictions into predictive distributions [ABSTRACT FROM AUTHOR]

This paper studies how streamflow predictability varies with basin characteristics. We introduce an index of basin complexity that is based on a model of least statistical complexity that is needed to reliably predict daily streamflow of the basin. We then relate it with climate, vegetation and soil characteristics of the basin. Daily streamflow is modeled using k nearest neighbor model of lagged streamflow that predicts next time step streamflow based on the occurrences of similar streamflow events from the past. In order to calculate basin complexity, we identify difficult streamflow events of the basin and then use Vapnik‐Chervonenkis generalization theory, which trades off model performance with Vapnik‐Chervonenkis dimension (i.e., a measure of model complexity), to find a k nearest neighbor model of appropriate complexity for predicting a difficult streamflow event of the basin. The average of selected model complexities corresponding to difficult events is then defined as the basin's complexity. Basin complexity of 412 Model Parameter Estimation Experiment basins from continental United States are then related with its six basin characteristics. All the characteristics have been derived from the Model Parameter Estimation Experiment database to represent climate, vegetation and soil characteristics of the basins in a concise manner. Results find that more complex basins that are drier have less seasonal rainfall, vegetation with more storage capacity (i.e., smaller 5‐week Normalized Difference Vegetation Index gradient), and faster responsive soils. The results reaffirm prior observations that minimum complexity that is required to model a basin depends on its climate and landscape characteristics (e.g., complex models do not perform well in dry basins). Key Points: Basin complexity is defined in terms of difficulty in learning statistical patterns of daily streamflowRelationship between complexity of MOPEX basins and hydrological characteristics is analyzedMore complex basins that are drier have less seasonal rainfall, vegetation with more storage capacity and faster responsive soils [ABSTRACT FROM AUTHOR]

Streamflow time series are commonly derived from stage‐discharge rating curves, but the uncertainty of the rating curve and resulting streamflow series are poorly understood. While different methods to quantify uncertainty in the stage‐discharge relationship exist, there is limited understanding of how uncertainty estimates differ between methods due to different assumptions and methodological choices. We compared uncertainty estimates and stage‐discharge rating curves from seven methods at three river locations of varying hydraulic complexity. Comparison of the estimated uncertainties revealed a wide range of estimates, particularly for high and low flows. At the simplest site on the Isère River (France), full width 95% uncertainties for the different methods ranged from 3 to 17% for median flows. In contrast, uncertainties were much higher and ranged from 41 to 200% for high flows in an extrapolated section of the rating curve at the Mahurangi River (New Zealand) and 28 to 101% for low flows at the Taf River (United Kingdom), where the hydraulic control is unstable at low flows. Differences between methods result from differences in the sources of uncertainty considered, differences in the handling of the time‐varying nature of rating curves, differences in the extent of hydraulic knowledge assumed, and differences in assumptions when extrapolating rating curves above or below the observed gaugings. Ultimately, the selection of an uncertainty method requires a match between user requirements and the assumptions made by the uncertainty method. Given the significant differences in uncertainty estimates between methods, we suggest that a clear statement of uncertainty assumptions be presented alongside streamflow uncertainty estimates. Plain Language Summary: Knowledge of the uncertainty in streamflow discharge measured at gauging stations is important for water management applications and scientific analysis. This paper shows that uncertainty estimates vary widely (typically up to a factor of 4) when comparing seven recently introduced estimation methods. A clear understanding of the assumptions underpinning different uncertainty estimation methods and the sources of uncertainty included in their calculations is needed when selecting a method and using and presenting its uncertainty estimates. Key Points: Methods for estimating the stage‐discharge rating curve and its uncertainty were compared for stream gauges with varying hydraulic complexityUncertainty estimates varied widely at high and low flows for the different methodsCareful description of the assumptions behind uncertainty methods is needed [ABSTRACT FROM AUTHOR]

While in catchment and hillslope hydrology a more nuanced approach is now taken to streamflow recession analysis, in the context of major aquifers it is commonly still assumed that the groundwater head recession rate will take exponential form, an idea originally proposed in the 19th Century. However it is shown here that, in early times, the groundwater head recession in a major aquifer should take an almost straight line form with a rate approximately equal to the long-term recharge rate divided by the aquifer storage coefficient. The length of this phase can be estimated from an analytical expression derived in the paper which depends on the aquifer diffusivity, length scale, and the position of the monitoring point. A transitional phase then leads to an exponential phase after some critical time which is independent of the position of the monitoring point. Major aquifers in a state of periodic quasi-steady state are expected to have rates of groundwater flux recession which deviate little from the average rate of groundwater recharge. Where quasi-exponential groundwater declines are observed in nature, their form may be diagnostic of particular types of aquifer properties and/or boundary effects, such as proximity to drainage boundaries, variations in transmissivity with hydraulic head, storage changes due to pumping, nonequilibrium flow at a range of spatial and temporal scales, and variations in specific yield with depth. Recession analysis has applicability to a range of groundwater problems and is powerful way of gaining insight into the hydrologic functioning of an aquifer. [ABSTRACT FROM AUTHOR]

This paper combines two approaches aimed at obtaining robust estimates of low-flow characteristics from short records and presents results for a low-flow index, the minimum monthly discharge with a 5 year return period (noted QMNA5). The two adjustment methods are first compared on a large set of catchments in France. The first method bases its adjustment on long-term climatic information and its sensitivity to the length of the short series is explored. The second method bases its adjustment on long-term streamflow information from a neighboring index catchment; its sensitivity to the proximity of the index catchment is analyzed. Finally, the climate-based and neighbor-based methods are combined, which confers specific robustness to the resulting flow adjustment approach. [ABSTRACT FROM AUTHOR]

The Turc-Mezentsev water balance formula is a simple and efficient mathematical formulation relating long-term average streamflow to long-term average precipitation and potential evaporation. Usually, it is applied in its original form, i.e., without parameterization. This paper aims to assess the possibilities of regionalizing this formula for improved efficiency on ungaged catchments. Four regionalization options are compared based on: (i) the relative error, (ii) the arithmetic difference, (iii) the correcting factor of the aridity index, and (iv) the shape factor. The four alternatives are assessed in terms of efficiency and robustness on a large set of 609 French catchments. We show that introducing a correcting factor of the aridity index provides the best regionalization solution. [ABSTRACT FROM AUTHOR]

A comparison of six procedures for estimating the linear trend parameter β in annual maximum 1 day river flows at five sites in southern Brazil showed marked differences between, on the one hand, estimates obtained by incorporating trend into the generalized extreme value (GEV) location parameter with all parameters estimated by maximum likelihood (ML) and on the other hand, estimates found by least squares, trend removal prior to fitting the GEV by ML, boot-strap sampling, and Theil-Sen estimation. ML estimates of trend were considerably smaller than those given by all other procedures. The same was true where trend had been incorporated into the Gumbel location parameter. Where 95% confidence intervals were calculated for the 'true' trend β by different procedures, some confidence intervals bracketed zero (indicating that the trend was not 'significant' at the 5% level), but there was no consistency between results from different procedures; Theil-Sen confidence intervals always bracketed zero, confidence intervals given by detrending never did. It is concluded that not only do different estimation procedures give different measures of trend uncertainty, as reported elsewhere, but the estimated trends themselves may differ, and the paper suggests an explanation why this may occur. Some philosophical issues relating to estimation of trend in climatological and hydrological extremes are discussed, and it is concluded that selection of a method to estimate trend must depend on context. [ABSTRACT FROM AUTHOR]

Stochastic dual dynamic programming (SDDP) is one of the few available algorithms to optimize the operating policies of large-scale hydropower systems. This paper presents a variant, called SDDPX, in which exogenous hydrologic variables, such as snow water equivalent and/or sea surface temperature, are included in the state space vector together with the traditional (endogenous) variables, i.e., past inflows. A reoptimization procedure is also proposed in which SDDPX-derived benefit-to-go functions are employed within a simulation carried out over the historical record of both the endogenous and exogenous hydrologic variables. In SDDPX, release policies are now a function of storages, past inflows, and relevant exogenous variables that potentially capture more complex hydrological processes than those found in traditional SDDP formulations. To illustrate the potential gain associated with the use of exogenous variables when operating a multireservoir system, the 3,137 MW hydropower system of Rio Tinto (RT) located in the Saguenay-Lac-St-Jean River Basin in Quebec (Canada) is used as a case study. The performance of the system is assessed for various combinations of hydrologic state variables, ranging from the simple lag-one autoregressive model to more complex formulations involving past inflows, snow water equivalent, and winter precipitation. [ABSTRACT FROM AUTHOR]

This paper investigates the impact of interevent variability of streamflow recession rates on two key streamflow statistics, namely, the equilibrium probability density function (pdf) and the autocorrelation. The relevance of the problem lies in the need to quantify and predict streamflow availability, on the basis of measurable rainfall and landscape attributes, by explicitly incorporating the stochasticity of the underlying climate and transport processes. Novel expressions for the seasonal pdf and the autocorrelation of the daily streamflows are derived by incorporating the randomness of the recession rates into a probabilistic framework where the streamflow fluctuations are explicitly related to the intermittency of the rainfall forcing. The presence of stochastic recession time constants is shown to impact only the third- and higher-order moments of the probability density function of the daily streamflows, without altering the mean and the variance of the pdf. A remarkable effect is instead produced on the correlation structure, which may exhibit long-term persistence even in the presence of a weak randomness of the recession rate. A relevant case study is then discussed to show that incorporating the stochasticity of the recession time constant leads to an improvement of the model ability to capture the behavior of the equilibrium streamflow pdf and of the underlying correlation function. [ABSTRACT FROM AUTHOR]

Daily streamflow time series are critical to a very broad range of hydrologic problems. Whereas daily streamflow time series are readily obtained from gaged catchments, streamflow information is commonly needed at catchments for which no measured streamflow information exists. At ungaged catchments, methods to estimate daily streamflow time series typically require the use of a reference streamgage, which transfers properties of the streamflow time series at a reference streamgage to the ungaged catchment. Therefore, the selection of a reference streamgage is one of the central challenges associated with estimation of daily streamflow at ungaged basins. The reference streamgage is typically selected by choosing the nearest streamgage; however, this paper shows that selection of the nearest streamgage does not provide a consistent selection criterion. We introduce a new method, termed the map-correlation method, which selects the reference streamgage whose daily streamflows are most correlated with an ungaged catchment. When applied to the estimation of daily streamflow at 28 streamgages across southern New England, daily streamflows estimated by a reference streamgage selected using the map-correlation method generally provides improved estimates of daily streamflow time series over streamflows estimated by the selection and use of the nearest streamgage. The map correlation method could have potential for many other applications including identifying redundancy and uniqueness in a streamgage network, calibration of rainfall runoff models at ungaged sites, as well as for use in catchment classification. [ABSTRACT FROM AUTHOR]

Research into the role of catchment vegetation within the hydrologic cycle has a long history in the hydrologic literature. Relationships between vegetation type and catchment evapotranspiration and runoff were primarily assessed through paired catchment studies during the 20th century. Results from over 200 paired catchment studies from around the world have been reported in the literature. Two constraints on utilizing the results from paired catchment studies in the wider domain have been that the catchment areas studied are generally (1) small (<10 km2) and (2) from a narrow range of climate types. The majority of reported paired catchment studies are located in the USA (∼47%) and Australia (∼27%) and experience mainly temperate (Köppen C) and cold (Köppen D) climate types. In this paper we assess the impact of vegetation type on mean annual evapotranspiration through a large, spatially, and climatically diverse data set of 699 catchments from around the world. These catchments are a subset of 861 unregulated catchments considered for the analysis. Spatially averaged precipitation and temperature data, in conjunction with runoff and land cover information, are analyzed to draw broad conclusions about the vegetation impact on mean annual evapotranspiration. In this analysis any vegetation impact signal is assessed through differences in long-term catchment average actual evapotranspiration, defined as precipitation minus runoff, between catchments grouped by vegetation type. This methodology differs from paired catchment studies where vegetation impact is assessed through streamflow responses to a controlled, within catchment, land cover change. The importance of taking the climate type experienced by the catchments into account when assessing the vegetation impact on evapotranspiration is demonstrated. Tropical and temperate forested catchments are found to have statistically significant higher median evapotranspiration, by about 170 mm and 130 mm, respectively, than non-forested catchments. Unexpectedly, cold forested catchments exhibit significantly lower median evapotranspiration, by about 90 mm, than non-forested catchments. No significant difference was found between median evapotranspiration of temperate evergreen and deciduous forested catchments though sample sizes were small. Temperate evergreen needleleaf forested catchments were found to have significantly higher median evapotranspiration than evergreen broadleaf forested catchments though sample sizes were small. The significant temperate forest versus non-forest difference in median evapotranspiration was found to persist for catchments with areas <1,000 km2, but not for catchments with areas ≥1,000 km2, which is consistent with the suggestion that the vegetation impact on evapotranspiration diminishes as catchment area increases. In summary, the results presented here are consistent with those drawn from reviews of paired catchment results. However, this paper demonstrates the value of a diverse hydroclimatic data set when assessing the vegetation impact on evapotranspiration as the magnitude of impact is observed to vary across climate types. [ABSTRACT FROM AUTHOR]

Nonlinear storage-discharge relations may explain important hydrologic features frequently observed under a variety of environmental conditions. In this paper the catchment dynamic storage problem is addressed by incorporating nonlinear storage-discharge relations in a stochastic framework to derive the statistical distribution and the duration curve of streamflows in river basins. Such a goal is achieved by extending recent analytical solutions for the seasonal probability distribution of discharges derived from the stochastic description of soil moisture dynamics at basin scales. Long-term probability density functions (pdf's) and flow duration curves are expressed through climatic, ecohydrologic, and geomorphic parameters of the basin by relaxing the linear reservoir assumption of subsurface flow components conveniently made in previous studies. In particular, this paper focuses on three different types of nonlinear, algebraic relationships between instantaneous discharges and storage volumes (concave and convex power laws and hyperbolic). Exact expressions of the streamflow pdf are derived in all cases, which are also tested by numerical simulations. Streamflow statistics and duration curves, estimates of catchment dynamic storages, recession timescales, and sensitivity to the underlying rainfall regime are then discussed. The model shows that different shapes of the streamflow pdf (bell shaped, monotonously decreasing, and bimodal) arise depending on the degree of nonlinearity of the storage-discharge relation, providing a new approach to catchment hydrology characterization. [ABSTRACT FROM AUTHOR]

Hydrological forecasts strongly rely on predictions of precipitation amounts and temperature as meteorological forcings for hydrological models. Ensemble weather predictions provide a number of different scenarios that reflect the uncertainty about these meteorological inputs, but these are often biased and under-dispersive, and therefore require statistical postprocessing. In addition to correcting the marginal distributions of the two weather variables, postprocessing methods must reconstruct their spatial, temporal, and intervariable dependence in order to generate physically realistic forecast trajectories that can be used as forcings of hydrological streamflow forecast models. For many years, a sample reordering method referred to as 'Schaake shuffle' has been used successfully to address this multivariate aspect of forecast distributions by using historical observation trajectories as multivariate 'dependence templates.' This paper proposes a variant of the Schaake shuffle, in which the historical dates are selected such that the marginal distributions of the corresponding observation trajectories are similar to the forecast marginal distributions, thus making it more likely that spatial and temporal gradients are preserved during the reordering procedure. This new approach is demonstrated with temperature and precipitation forecasts over four river basins in California, and it is shown to improve upon the standard Schaake shuffle both with respect to verification metrics applied to the forcings, and verification metrics applied to the resulting streamflow predictions. [ABSTRACT FROM AUTHOR]

The paper 'Climate and agricultural land use change impacts on streamflow in the upper midwestern United States' by Satish C. Gupta, Andrew C. Kessler, Melinda K. Brown, and Francis Zvomuya (hereafter referred to as Gupta et al.) purports to evaluate 'the relative importance of changes in precipitation and LULC (land use, land cover) on streamflow in 29 Hydrologic Unit Code 008 watersheds in the Upper Midwestern United States.' However, as we report here, the approach used by Gupta et al. is wholly inadequate for making such an evaluation. Gupta et al. use strong language to criticize other studies and imply a level of certainty that goes well beyond, and in some cases is entirely unsupported by, the results they have presented. We take this opportunity to point out several critical flaws in their study. [ABSTRACT FROM AUTHOR]

The aim of this paper is to describe and evaluate a hybrid spectral- and time-domain approach for the calibration of shot noise models for daily streamflow generation. The calibration approach allows the parameter estimation of a minimum-phase rainfall/streamflow model using two steps. In the predictor step, the power spectral density of a recorded streamflow series is used to calibrate the parameters connected with the model dynamics. During the corrector step, a classic time-domain procedure is used to calibrate the parameters connected with the average output of the model and the parameters that characterize the rainfall stochastic process. The procedure is demonstrated through its application to the daily streamflow time series associated with three Italian watersheds, and its results are then compared with those obtained by means of a time-domain calibration method available in the literature. [ABSTRACT FROM AUTHOR]

Recursive digital filters (RDFs) are widely used for estimating base flow from streamflow hydrographs, and various forms of RDFs have been developed based on different physical models. Numerical experiments have been used to objectively evaluate their performance, but they have not been sufficiently comprehensive to assess a wide range of RDFs. This paper extends these studies to understand the limitations of a generalized RDF method as a pathway for future field calibration. Two formalisms are presented to generalize most existing RDFs, allowing systematic tuning of their complexity. The RDFs with variable complexity are evaluated collectively in a synthetic setting, using modeled daily base flow produced by Li et al. (2014) from a range of synthetic catchments simulated with HydroGeoSphere. Our evaluation reveals that there are optimal RDF complexities in reproducing base flow simulations but shows that there is an inherent physical inconsistency within the RDF construction. Even under the idealized setting where true base flow data are available to calibrate the RDFs, there is persistent disagreement between true and estimated base flow over catchments with small base flow components, low saturated hydraulic conductivity of the soil and larger surface runoff. The simplest explanation is that low base flow 'signal' in the streamflow data is hard to distinguish, although more complex RDFs can improve upon the simpler Eckhardt filter at these catchments. [ABSTRACT FROM AUTHOR]

This paper examines the extent to which regime-like behavior in streamflow time series impacts reservoir operating policy performance. We begin by incorporating a regime state variable into a well-established stochastic dynamic programming model. We then simulate and compare optimized release policies-with and without the regime state variable-to understand how regime shifts affect operating performance in terms of meeting water delivery targets. Our optimization approach uses a Hidden Markov Model to partition the streamflow time series into a small number of separate regime states. The streamflow persistence structures associated with each state define separate month-to-month streamflow transition probability matrices for computing penalty cost expectations within the optimization procedure. The algorithm generates a four-dimensional array of release decisions conditioned on the within-year time period, reservoir storage state, inflow class, and underlying regime state. Our computational experiment is executed on 99 distinct, hypothetical water supply reservoirs fashioned from the Australian Bureau of Meteorology's Hydrologic Reference Stations. Results show that regime-like behavior is a major cause of suboptimal operations in water supply reservoirs; conventional techniques for optimal policy design may misguide the operator, particularly in regions susceptible to multiyear drought. Stationary streamflow models that allow for regime-like behavior can be incorporated into traditional stochastic optimization models to enhance the flexibility of operations. [ABSTRACT FROM AUTHOR]