Predictions In Ungauged Basins Research Articles

Predicting the ungauged basin: model validation and realism assessment

The hydrological decade on Predictions in Ungauged Basins (PUB) led to many new insights in model development, calibration strategies, data acquisition and uncertainty analysis. Due to a limited amount of published studies on genuinely ungauged basins, model validation and realism assessment of model outcome has not been discussed to a great extent. With this paper we aim to contribute to the discussion on how one can determine the value and validity of a hydrological model developed for an ungauged basin. As in many cases no local, or even regional, data are available, alternative methods should be applied. Using a PUB case study in a genuinely ungauged basin in southern Cambodia, we give several examples of how one can use different types of soft data to improve model design, calibrate and validate the model, and assess the realism of the model output. A rainfall-runoff model was coupled to an irrigation reservoir, allowing the use of additional and unconventional data. The model was mainly forced with remote sensing data, and local knowledge was used to constrain the parameters. Model realism assessment was done using data from surveys. This resulted in a successful reconstruction of the reservoir dynamics, and revealed the different hydrological characteristics of the two topographical classes. This paper does not present a generic approach that can be transferred to other ungauged catchments, but it aims to show how clever model design and alternative data acquisition can result in a valuable hydrological model for an ungauged catchment.

Stream temperature prediction in ungauged basins: review of recent approaches and description of a new physics-derived statistical model

Abstract. The development of stream temperature regression models at regional scales has regained some popularity over the past years. These models are used to predict stream temperature in ungauged catchments to assess the impact of human activities or climate change on riverine fauna over large spatial areas. A comprehensive literature review presented in this study shows that the temperature metrics predicted by the majority of models correspond to yearly aggregates, such as the popular annual maximum weekly mean temperature (MWMT). As a consequence, current models are often unable to predict the annual cycle of stream temperature, nor can the majority of them forecast the inter-annual variation of stream temperature. This study presents a new statistical model to estimate the monthly mean stream temperature of ungauged rivers over multiple years in an Alpine country (Switzerland). Contrary to similar models developed to date, which are mostly based on standard regression approaches, this one attempts to incorporate physical aspects into its structure. It is based on the analytical solution to a simplified version of the energy-balance equation over an entire stream network. Some terms of this solution cannot be readily evaluated at the regional scale due to the lack of appropriate data, and are therefore approximated using classical statistical techniques. This physics-inspired approach presents some advantages: (1) the main model structure is directly obtained from first principles, (2) the spatial extent over which the predictor variables are averaged naturally arises during model development, and (3) most of the regression coefficients can be interpreted from a physical point of view – their values can therefore be constrained to remain within plausible bounds. The evaluation of the model over a new freely available data set shows that the monthly mean stream temperature curve can be reproduced with a root-mean-square error (RMSE) of ±1.3 °C, which is similar in precision to the predictions obtained with a multi-linear regression model. We illustrate through a simple example how the physical aspects contained in the model structure can be used to gain more insight into the stream temperature dynamics at regional scales.

Analyzing the effects of excess rainfall properties on the scaling structure of peak discharges: Insights from a mesoscale river basin

AbstractKey theoretical and empirical results from the past two decades have established that peak discharges resulting from a single rainfall‐runoff event in a nested watershed exhibit a power law, or scaling, relation to drainage area and that the parameters of the power law relation, henceforth referred to as the flood scaling exponent and intercept, change from event to event. To date, only two studies have been conducted using empirical data, both using data from the 21 km2 Goodwin Creek Experimental Watershed that is located in Mississippi, in an effort to uncover the physical processes that control the event‐to‐event variability of the flood scaling parameters. Our study expands the analysis to the mesoscale Iowa River basin (A = 32,400 km2), which is located in eastern Iowa, and provides additional insights into the physical processes that control the flood scaling parameters. Using 51 rainfall‐runoff events that we identified over the 12 year period since 2002, we show how the duration and depth of excess rainfall, which is the portion of rainfall that contributes to direct runoff, control the flood scaling exponent and intercept. Moreover, using a diagnostic simulation study that is guided by evidence found in empirical data, we show that the temporal structure of excess rainfall has a significant effect on the scaling structure of peak discharges. These insights will contribute toward ongoing efforts to provide a framework for flood prediction in ungauged basins.

Evaluation of Regionalization Methods for Hourly Continuous Streamflow Simulation Using Distributed Models in Boreal Catchments

AbstractRegionalization for prediction in ungauged basins at hourly resolution is important for water resources management (e.g., floods and hydropeaking). In the research reported in this paper, calibration of 26 catchments (39–3,090 km2) in mid-Norway was performed using hourly records and three spatially distributed (1×1 km2) precipitation–runoff models, as follows: (1) first-order nonlinear system model, (2) Hydrologiska Byrans Vattenballansavdelning (HBV) model, and (3) basic grid model. Four regionalization methods for each model [(1) parameter set yielding maximum regional weighted average performance measures (PMs), (2) regional median of optimal parameters, (3) nearest neighbor (NN), and (4) physical similarity (PS)] were evaluated and compared with three benchmarks. Parameter transfer from best regional donor and from an ideal best arbitrary single-donor, and local calibration (LC) were as benchmarks. The PS attributes include hypsometric curves, land use, drainage density, catchment area, ter...

Panta Rhei, the new science decade of IAHS

Abstract. To address the water related problems that society is facing in this time and age, hydrologists can no longer consider the hydrological system as if there is no human influence. The human and water systems are closely intertwined and in this interaction patterns emerge that are the result of two co-evolving systems. The hydrology and the water resources form conditions under which societies develop, but as a result, and in turn, societies influence the functioning of the hydrological system. This calls for a new science that tries to understand these interactions with the objective to explore feasible futures and strategies to guide this co-evolution towards sustainable use of our natural resources. After the closure of the PUB decade on Prediction in Ungauged Basins, the International Association of Hydrological Sciences (IAHS) has taken up the challenge of trying to make fundamental contributions to this new science of integrated hydrological and societal processes under the title of "Panta Rhei", symbolising that everything changes in a perpetual dynamic interaction between people and water. This may be a big step in the science of hydrology, venturing well outside the comfortable boundaries of pristine hydrology, but the urgency of water issues around the globe forces us to do so, and in the process, hydrologists are likely to discover new paradigms, new modelling approaches, enhanced understanding of the human-water system, and hopefully new ways to steer our society towards a more sustainable future.

Prediction in ungauged estuaries: An integrated theory

AbstractMany estuaries in the world are ungauged. The International Association of Hydrological Sciences completed its science decade on Prediction in Ungauged Basins (PUB) in 2012 (Hrachowitz et al., ). Prediction on the basis of limited data is a challenge in hydrology, but not less so in estuaries, where data on fundamental processes are often lacking. In this paper, relatively simple, but science‐based, methods are presented that allow researchers, engineers, and water managers to obtain first‐order estimates of essential process parameters in estuaries, such as the estuary depth, the tidal amplitude, the tidal excursion, the phase lag, and the salt water intrusion, on the basis of readily obtainable information, such as topographical maps and tidal tables. These apparently simple relationships are assumed to result from the capacity of freely erodible water bodies to adjust themselves to external drivers and to dissipate the free energy from these drivers as efficiently as possible. Thus, it is assumed that these systems operate close to their thermodynamic limit, resulting in predictable patterns that can be described by relatively simple equations. Although still much has to be done to develop an overall physics‐based theory, this does not prevent us from making use of the empirical “laws” that we observe in alluvial estuaries.

A Landscape‐Based Regionalization of Natural Flow Regimes in the Ebro River Basin and Its Biological Validation

AbstractFlow prediction in ungauged basins is an important task for water resources planning and management, and remains a fundamental challenge for hydroecological research. Based on a previous classification of streams and rivers in the Ebro River basin (Spain), where six natural flow‐regime types were identified, we apply a new predictive approach of the flow regime type based on climatic and physiographic descriptors. We used a set of easily available environmental variables as discriminant parameters: annual precipitation, annual evapotranspiration, annual air temperature, elevation, catchment area, drainage density and geology. A stepwise landscape‐based classification procedure consisting of several stepwise discriminant analyses and canonical discriminant analyses allocated a set of sites with poor or no natural flow data into the flow types defined. Misclassification rates obtained by cross‐validation ranged between 1.12% and 11.9%. Additionally, the ecological soundness of the proposed regionalization was tested by the concordance between macroinvertebrate communities and the proposed classification using NMDS and ANOSIM. NMDS resulted in a clear separation of sites into five NFR classes with available macroinvertebrate data, and ANOSIM found significant differences in macroinvertebrate communities among classes. Copyright © 2014 John Wiley & Sons, Ltd.

A constraint-based search algorithm for parameter identification of environmental models

Abstract. Many environmental systems models, such as conceptual rainfall-runoff models, rely on model calibration for parameter identification. For this, an observed output time series (such as runoff) is needed, but frequently not available (e.g., when making predictions in ungauged basins). In this study, we provide an alternative approach for parameter identification using constraints based on two types of restrictions derived from prior (or expert) knowledge. The first, called parameter constraints, restricts the solution space based on realistic relationships that must hold between the different model parameters while the second, called process constraints requires that additional realism relationships between the fluxes and state variables must be satisfied. Specifically, we propose a search algorithm for finding parameter sets that simultaneously satisfy such constraints, based on stepwise sampling of the parameter space. Such parameter sets have the desirable property of being consistent with the modeler's intuition of how the catchment functions, and can (if necessary) serve as prior information for further investigations by reducing the prior uncertainties associated with both calibration and prediction.

An experiment to gauge an ungauged catchment: rapid data assessment and eco-hydrological modelling in a data-scarce rural catchment

We conducted a PUB (predictions in ungauged basins) experiment looking at hydrology and crop dynamics in the semi-arid rural Mod catchment in India. The experiment was motivated by the aims (a) to develop a coupled eco-hydrological model capable of analysing land-use strategies concerning crop water need, erosion protection, crop yield and resistivity against droughts and floods, and (b) to assess the feasibility of a strategy for collecting the necessary data in a data-scarce region. Our experiment combines parsimonious data assessment and eco-hydrological model coupling at the lower mesoscale. Linking bottom-up sampling of functionally representative soil classes and top-down regionalization based on spectral properties of the same resulted in a comprehensive distributed data basis for the model. A clear focus on the dominating processes and the catena as the organizing landscape element in the given environmental setting enabled this. We employed the WASA (Water Availability in Semi-Arid environments) model for uncalibrated process-based water balance modelling and integrated a crop simulation subroutine based on the SWAP (Soil Water Atmosphere Plant) model to account for crop dynamics, feedbacks and yield estimation. While we found the data assessment strategy and the hydrological model application largely feasible, in terms of its accounting for scale, processes and model concepts, the simulation of feedbacks with crops was problematic. Contributing to the PUB issue, more general conclusions are drawn concerning spatially-distributed structural information and uncalibrated modelling.Editor Z.W. Kundzewicz; Associate editor F. Hattermann

On a PUB methodology from Chinese lessons

Based on a Chinese saying: “Even a clever housewife cannot cook a meal without rice”, a simple categorization of the methods for Predictions in Ungauged Basins (PUB) is proposed, including: Borrowing, obtaining hydrological information by transplanting measurements from a similar basin, or extrapolating/interpolating the data from neighbouring catchments; Substituting, finding substitutes either from the ungauged basin or from donating area(s); and Generating, obtaining data via field or laboratory observations. The Substituting category is classified further into: S1, substitution only from within the ungauged basin using fully process-based models without calibration; S2-1, from similar gauged basins using established index/distribution; S2-2, from various gauged basins using regression and/or process-based relationships between the climate/catchment features and hydrological signatures (CCH), and S3, from the information beyond the CCH relationship. Based on a review, the Darwinian S2-2 and Newtonian S1 were found to be the two most popular methods, both for China and worldwide PUB.Editor Z.W. Kundzewicz

Regionalization of hydrologic response in the Great Lakes basin: Considerations of temporal scales of analysis

Summary Methods for predicting streamflow in areas with limited or nonexistent measures of hydrologic response commonly rely on regionalization techniques, where knowledge pertaining to gauged watersheds is transferred to ungauged watersheds. Hydrologic response indices have frequently been employed in contemporary regionalization research related to predictions in ungauged basins. In this study, we developed regionalization models using multiple linear regression and regression tree analysis to derive relationships between hydrologic response and watershed physical characteristics for 163 watersheds in the Great Lakes basin. These models provide an empirical means for simulating runoff in ungauged basins at a monthly time step without implementation of a rainfall–runoff model. For the dependent variable in these regression models, we used monthly runoff ratio as the indicator of hydrologic response and defined it at two temporal scales: (1) treating all monthly runoff ratios as individual observations, and (2) using the mean of these monthly runoff ratios for each watershed as a representative observation. Application of the models to 62 validation watersheds throughout the Great Lakes basin indicated that model simulations were far more sensitive to the temporal characterization of hydrologic response than to the type of regression technique employed, and that models conditioned on individual monthly runoff ratios (rather than long term mean values) performed better. This finding is important in light of the increased usage of hydrologic response indices in recent regionalization studies. Models using individual observations for the dependent variable generally simulated monthly runoff with reasonable skill in the validation watersheds (median Nash–Sutcliffe efficiency = 0.53, median R 2 = 0.66, median magnitude of the deviation of runoff volume = 13%). These results suggest the viability of empirical approaches to simulate runoff in ungauged basins. This finding is significant given the many regions of the world with sparse gauging networks and limited resources for gathering the field data required to calibrate rainfall–runoff models.

Use of a parsimonious rainfall–run‐off model for predicting hydrological response in ungauged basins

AbstractA new parameter parsimonious rainfall–run‐off model, the Distance Distribution Dynamics (DDD) model, is used to simulate hydrological time series at ungauged sites in the Lygne basin in Norway. The model parameters were estimated as functions of catchment characteristics determined by geographical information system. The multiple regression equations relating catchment characteristics and model parameters were trained from 84 calibrated catchments located all over Norway, and all model parameters showed significant correlations with catchment characteristics. The significant correlation coefficients (with p‐value < 0.05) ranged from 0.22 to 0.55. The suitability of DDD for predictions in ungauged basins was tested for 17 catchments not used to estimate the multiple regression equations. For ten of the 17 catchments, deviations in Nash–Sutcliffe efficiency (NSE) criteria between the calibrated and regionalised model were less than 0.1, and for two calibrated catchments within the Lygne basin, the deviations were less than 0.08. The median NSE for the regionalized DDD for the 17 catchments for two time series was 0.66 and 0.72. Deviations in NSE between calibrated and regionalised models are well explained by the deviations between calibrated and regressed parameters describing spatial snow distribution and snowmelt respectively. The quality of the simulated run‐off series for the ungauged sites in the Lygne basin was assessed by comparing flow indices describing high, medium and low flow estimated from observed run‐off at the 17 catchments and for the simulated run‐off series. The indices estimated for the simulated series were generally well within the ranges defined by the 17 observed series. Copyright © 2014 John Wiley & Sons, Ltd.

Regionalization of subsurface stormflow parameters of hydrologic models: Derivation from regional analysis of streamflow recession curves

Summary Subsurface stormflow is an important component of the rainfall–runoff response, especially in steep terrain. Its contribution to total runoff is, however, poorly represented in the current generation of land surface models. The lack of physical basis of these common parameterizations precludes a priori estimation of the stormflow (i.e. without calibration), which is a major drawback for prediction in ungauged basins, or for use in global land surface models. This paper is aimed at deriving regionalized parameterizations of the storage–discharge relationship relating to subsurface stormflow from a top–down empirical data analysis of streamflow recession curves extracted from 50 eastern United States catchments. Detailed regression analyses were performed between parameters of the empirical storage–discharge relationships and the controlling climate, soil and topographic characteristics. The regression analyses performed on empirical recession curves at catchment scale indicated that the coefficient of the power-law form storage–discharge relationship is closely related to the catchment hydrologic characteristics, which is consistent with the hydraulic theory derived mainly at the hillslope scale. As for the exponent, besides the role of field scale soil hydraulic properties as suggested by hydraulic theory, it is found to be more strongly affected by climate (aridity) at the catchment scale. At a fundamental level these results point to the need for more detailed exploration of the co-dependence of soil, vegetation and topography with climate.

Regionalization of subsurface stormflow parameters of hydrologic models: Up-scaling from physically based numerical simulations at hillslope scale

Summary Subsurface stormflow is an important component of the rainfall-runoff response, especially in steep forested regions. However; its contribution is poorly represented in current generation of land surface hydrological models (LSMs) and catchment-scale rainfall-runoff models. The lack of physical basis of common parameterizations precludes a priori estimation (i.e. without calibration), which is a major drawback for prediction in ungauged basins, or for use in global models. This paper is aimed at deriving physically based parameterizations of the storage-discharge relationship relating to subsurface flow. These parameterizations are derived through a two-step up-scaling procedure: firstly, through simulations with a physically based (Darcian) subsurface flow model for idealized three dimensional rectangular hillslopes, accounting for within-hillslope random heterogeneity of soil hydraulic properties, and secondly, through subsequent up-scaling to the catchment scale by accounting for between-hillslope and within-catchment heterogeneity of topographic features (e.g., slope). These theoretical simulation results produced parameterizations of the storage-discharge relationship in terms of soil hydraulic properties, topographic slope and their heterogeneities, which were consistent with results of previous studies. Yet, regionalization of the resulting storage-discharge relations across 50 actual catchments in eastern United States, and a comparison of the regionalized results with equivalent empirical results obtained on the basis of analysis of observed streamflow recession curves, revealed a systematic inconsistency. It was found that the difference between the theoretical and empirically derived results could be explained, to first order, by climate in the form of climatic aridity index. This suggests a possible co-dependence of climate, soils, vegetation and topographic properties, and suggests that subsurface flow parameterization needed for ungauged locations must account for both the physics of flow in heterogeneous landscapes, and the co-dependence of soil and topographic properties with climate, including possibly the mediating role of vegetation.

Continuous streamflow prediction in ungauged basins: The effects of equifinality and parameter set selection on uncertainty in regionalization approaches

AbstractThis 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.

Uncertainty analysis in model parameters regionalization: a case study involving the SWAT model in Mediterranean catchments (Southern France)

Abstract. In this study a method for propagating the hydrological model uncertainty in discharge predictions of ungauged Mediterranean catchments using a model parameter regionalization approach is presented. The method is developed and tested for the Thau catchment located in Southern France using the SWAT hydrological model. Regionalization of model parameters, based on physical similarity measured between gauged and ungauged catchment attributes, is a popular methodology for discharge prediction in ungauged basins, but it is often confronted with an arbitrary criterion for selecting the "behavioral" model parameter sets (Mps) at the gauged catchment. A more objective method is provided in this paper where the transferrable Mps are selected based on the similarity between the donor and the receptor catchments. In addition, the method allows propagating the modeling uncertainty while transferring the Mps to the ungauged catchments. Results indicate that physically similar catchments located within the same geographic and climatic region may exhibit similar hydrological behavior and can also be affected by similar model prediction uncertainty. Furthermore, the results suggest that model prediction uncertainty at the ungauged catchment increases as the dissimilarity between the donor and the receptor catchments increases. The methodology presented in this paper can be replicated and used in regionalization of any hydrological model parameters for estimating streamflow at ungauged catchment.

Special Issue on Grand Challenges in Hydrology

There is no question that we have come a long way in hydrology, but we may just be in the beginning yet. This paradox can be justified on many grounds. For instance: (1) while hydrology as the “study of water” is as old as the human civilization, it is still in an early phase of being established as a science in its own right; (2) although we have always studied water with respect to its surroundings, our understanding of the actual interactions between hydrologic systems and other natural/physical systems (e.g., climate, environmental, biological, ecological) is still in a state of infancy; and (3) while water has always been an integral part of humans, we are only now starting to study the influence of humans on water and vice versa, especially under the framework of coupled human-water relations. The state of hydrology, especially its future, is further complicated by the uncertainties associated with (1) population growth and exploitation/alteration of water and other natural resources, especially in the developing and least-developed regions where such resources are already scarce; and (2) impacts of global climate change on our natural and social systems at global, regional, and local levels. Under these circumstances, a fundamental requirement toward true progress in hydrology is to identify the most challenging problems (or “grand challenges”) and particularly focus our studies on such. The grand challenges are defined by the depth of their scientific meanings and/or the breadth of their societal implications. While there is no doubt that we have made and continue to make efforts in this direction, what is also clear is that such efforts have unfortunately been far and few between. There is, therefore, a need to properly address this deficiency, which provides the motivation for the current special issue on Grand Challenges in Hydrology. This special issue compiles articles that address some of the grand challenges in hydrology that require urgent attention. The contributors come from over 30 universities and research institutions across 11 countries around the world. The articles cover a wide range of topics and associated challenges: parameter estimation in hydrologic models and data issues—focusing on information content (Beven and Smith), resampling method (Sikorska et al.), and predictions in ungauged basins (Seibert and McDonnell); catchment classification (Sivakumar et al.); climate change impact studies—freshwater resources (Kundzewicz and Gerten), sea-level rise (Chang et al.), and downscaling method (Jang and Kavvas); groundwater—modeling and management (Ojha et al.) and water quality (Ramadas et al.); hydrosystemecosystem interactions (Zalewski); dams and development (Tortajada); and hydrology education (Ruddell and Wagener). Each article roughly qualifies as a review on the topic discussed, offering a comprehensive account of the current status, unresolved issues, and important directions for the future. We, the guest editors, clearly recognize that the list of grand challenges compiled in this special issue remains incomplete because there are indeed many other topics that deserve serious consideration. Examples of such topics are (1) hydrologic measurement techniques; (2) emerging techniques for analysis of hydrologic systems (e.g., hydroinformatics); (3) land-use change hydrology; (4) surface water–groundwater interactions; (5) global water scarcity issues and solutions; (6) coupled human–water relations; and (7) transboundary water management (including the role of hydropolitics). The selection of the topics and challenges for this special issue was influenced by the following, among others: identification of all the grand challenges was a challenge in itself; and time constraints, especially faced by potential contributors of some topics. Nevertheless, we believe that this special issue is a very useful contribution to the discussion of grand challenges in particular and to the advancement of hydrology broadly.

Flood design recipes vs. reality: can predictions for ungauged basins be trusted?

Abstract. Despite the great scientific and technological advances in flood hydrology, everyday engineering practices still follow simplistic approaches that are easy to formally implement in ungauged areas. In general, these "recipes" have been developed many decades ago, based on field data from typically few experimental catchments. However, many of them have been neither updated nor validated across all hydroclimatic and geomorphological conditions. This has an obvious impact on the quality and reliability of hydrological studies, and, consequently, on the safety and cost of the related flood protection works. Preliminary results, based on historical flood data from Cyprus and Greece, indicate that a substantial revision of many aspects of flood engineering procedures is required, including the regionalization formulas as well as the modelling concepts themselves. In order to provide a consistent design framework and to ensure realistic predictions of the flood risk (a key issue of the 2007/60/EU Directive) in ungauged basins, it is necessary to rethink the current engineering practices. In this vein, the collection of reliable hydrological data would be essential for re-evaluating the existing "recipes", taking into account local peculiarities, and for updating the modelling methodologies as needed.

Assessing hydrologic prediction uncertainty resulting from soft land cover classification

Summary For predictions in ungauged basins (PUB), environmental data is generally not available and needs to be inferred by indirect means. Existing technologies such as remote sensing are valuable tools for estimating the lacking data, as these technologies become more widely available and have a high areal coverage. However, indirect estimates of the environmental characteristics are prone to uncertainty. Hence, an improved understanding of the quality of the estimates and the development of methods for dealing with their associated uncertainty are essential to evolve towards accurate PUB. In this study, the impact of the uncertainty associated with the classification of land cover based on multi-temporal SPOT imagery, resulting from the use of the Random Forests classifier, on the predictions of the hydrologic model TOPLATS is investigated through a Monte Carlo simulation. The results show that the predictions of evapotranspiration, runoff and baseflow are hardly affected by the classification uncertainty when area-averaged predictions are intended, implying that uncertainty propagation is only advisable in case a spatial distribution of the predictions is relevant for decision making or is coupled to other spatially distributed models. Based on the resulting uncertainty map, guidelines for additional data collection are formulated in order to reduce the uncertainty for future model applications. Because a Monte Carlo-based uncertainty analysis is computationally very demanding, especially when complex models are involved, we developed a fast indicative uncertainty assessment method that allows for generating proxies of the Monte Carlo-based result in terms of the mean prediction and its associated uncertainty based on a single model evaluation. These proxies are shown to perform well and provide a good indication of the impact of classification uncertainty on the prediction result.

A Distributed Hydrological Model Driven by Multi-Source Spatial Data and Its Application in the Ili River Basin of Central Asia

Hydrological simulation in ungauged regions is a popular topic in water resource and environmental research, and is also an important part of the international research initiative Predictions in Ungauged Basins (PUB). In this study, a multi-spatial data-based DTVGM (MS-DTVGM), combining multi-source spatial data (MS-spatial data) with the Distributed Time-Variant Gain Model (DTVGM), was built in order to reduce dependence on conventional observation, and was applied to the Ili River basin where traditional data sets are scarce. Because it utilizes MS-spatial data to measure precipitation, potential evapotranspiration, air temperature, vegetation parameters, and soil parameters, the model is driven purely by data from common platforms, thus overcoming the disadvantage of the large amounts of data typically required for distributed hydrological models. The inputs and simulation results were calibrated and validated using station or field observations. The results indicate that: 1) the MS-DTVGM is feasible in the Ili River basin; all model inputs can be acquired from multi-source spatial data and the key parameters are accurate; 2) the MS-DTVGM has good performance on a monthly time scale, and its simulation results can be used for a longer time-scale water resource analysis; and (3) daily runoff generation correlated strongly with snowmelt, the R2 is about 0.69 indicating that the latter is an important contributor to water resources and suggesting that a snowmelt module is indispensable this area. The potential of distributed models for hydrological simulation in data-scarce regions using MS-spatial data was clearly demonstrated.