Search

In mountainous areas, accurately estimating the long‐term climatology of seasonal precipitations is challenging due to the lack of high‐altitude rain gauges and the complexity of the topography. This study addresses these challenges by interpolating seasonal precipitation data from 3189 rain gauges across France over the 1982–2018 period, using geographical coordinates, and altitude. In this study, an additional predictor is provided from simulations of a Convection‐Permitting Regional Climate Model (CP‐RCM). The simulations are averaged to obtain seasonal precipitation climatology, which helps capture the relationship between topography and long‐term seasonal precipitation. Geostatistical and machine learning models are evaluated within a cross‐validation framework to determine the most appropriate approach to generate seasonal precipitation reference fields. Results indicate that the best model uses a machine learning approach to interpolate the ratio between long‐term seasonal precipitation from observations and CP‐RCM simulations. This method successfully reproduces both the mean and variance of observed data, and slightly outperforms the best geostatistical model. Moreover, incorporating the CP‐RCM outputs as an explanatory variable significantly improves interpolation accuracy and altitude extrapolation, especially when the rain gauge density is low. These results imply that the commonly used altitude‐precipitation relationship may be insufficient to derive seasonal precipitation fields. The CP‐RCM simulations, increasingly available worldwide, present an opportunity for improving precipitation interpolation, especially in sparse and complex topographical regions. [ABSTRACT FROM AUTHOR]

The effects of spatial aggregation on the skill of downscaled precipitation predictions obtained over an 8 × 8 km2 grid from circulation analogs for metropolitan France are explored. The Safran precipitation reanalysis and an analog approach are used to downscale the precipitation where the predictors are taken from the 40-yr ECMWF Re-Analysis (ERA-40). Prediction skill-characterized by the continuous ranked probability score (CRPS), its skill score, and its decomposition-is generally found to continuously increase with spatial aggregation. The increase is also greater when the spatial correlation of precipitation is lower. This effect is shown from an empirical experiment carried out with a fully uncorrelated dataset, generated from a space-shake experiment, where the precipitation time series of each grid cell is randomly assigned to another grid cell. The underlying mechanisms of this effect are further highlighted with synthetic predictions simulated using a stochastic spatiotemporal generator. It is shown 1) that the skill increase with spatial aggregation jointly results from the higher and lower values obtained for the resolution and uncertainty terms of the CRPS decomposition, respectively, and 2) that the lower spatial correlation of precipitation is beneficial for both terms. Results obtained for France suggest that the prediction skill indefinitely increases with aggregation. A last experiment is finally proposed to show that this is not expected to be always the case. A prediction skill optimum is, for instance, obtained when the mean areal precipitation is estimated over a region where local precipitations of different grid cells originate from different underlying meteorological processes. [ABSTRACT FROM AUTHOR]

As a consequence of the increase in atmospheric greenhouse gas concentrations, potential changes in both precipitation occurrence and intensity may lead to several consequences for Earth's environment. It is therefore relevant to estimate these changes in order to anticipate their consequences. Many studies have been published on precipitation changes based on climate simulations. These studies are almost always based on time slices; precipitation changes are estimated by comparing two 30-yr windows. To this extent, it is commonly assumed that nonstationary processes are not significant for such a 30-yr slice. Thus, it frees the investigator to statistically model nonstationary processes. However, using transient runs instead of time slices surely leads to more accurate analysis since more data are taken into account. Therefore, the aim of the present study was to develop a transient probabilistic model for describing simulated daily precipitation from the Canadian Regional Climate Model (CRCM) in order to investigate precipitation evolution over North America. Changes to both the occurrence and intensity of precipitation are then assessed from a continuous time period. Extreme values are also investigated with the transient run; a new methodology using the models for precipitation occurrence and intensity was developed for achieving nonstationary frequency analysis. The results herein show an increase in both precipitation occurrence and intensity for most parts of Canada while a decrease is expected over Mexico. For the continental United States, a decrease in both occurrence and intensity is expected in summer but an increase is expected in winter. [ABSTRACT FROM AUTHOR]

Highlights: [•] The dependence between peak and volume of flow of is modelled with a copula family. [•] Parameter joint inference and model selection are performed using a Bayesian method. [•] Cross correlation between margin parameters are stronger than expected. [•] Marginal distributions cannot be forgotten in the model selection process. [•] Special attention must be addressed to model validation in a context of extremes. [ABSTRACT FROM AUTHOR]

Stochastic point processes for rainfall are known to be able to preserve the temporal variability of rainfall on several levels of aggregation (e.g. hourly, daily), especially when the cluster approach is used. One major assumption in most of the applications todate is the stationarity of the rainfall properties in time, which must be reconsidered under a climate change hypothesis. Here, we propose new theoretical developments of a Poisson-based model with cluster, namely the Neyman-Scott Rectangular Pulses Model, which treats storm frequency as a nonstationary function. In this paper, storm frequency is modelled as a linear function of time in order to reproduce an increase (or decrease) of the mean annual precipitation. The basic theory is reconsidered and the moments are derived up to the third order. Then, a calibration method based on the generalized method of moments is proposed and discussed. An application to a rainfall time series from Uccle illustrates how this model can reproduce a trend for the average rainfall. This work opens new avenues for future developments on transient stochastic rainfall models and highlights the major challenges linked to this approach. [ABSTRACT FROM AUTHOR]

Real-time assessments of stream flows under ice cover, using two distinct objective approaches, were compared over a 5-year period. Approaches based on artificial neural networks, defining mathematical relationships between stream flow, water level, and air temperature, and on a deterministic hydrological model were applied at eight gauged sites located in southern Quebec. Good results were obtained using both approaches, when no snowmelt contributes to the rise of the inflows. In the other hydrological situations, the neural network results were the best, but results of both approaches were sensibly poorer. Nevertheless, the potential for increasing the skills of the deterministic model seems high. Otherwise, a preliminary analysis showed that both approaches lead to stream-flow estimations that are not radically worse than the ones performed by the team of experts. [ABSTRACT FROM AUTHOR]

In official statistics, when a file of microdata must be delivered to external users, it is very difficult to propose them a file where missing values has been treated by multiple imputations. In order to overcome this difficulty, we propose a method of single imputation for qualitative data that respect numerous constraints. The imputation is balanced on totals previously estimated; editing rules can be respected; the imputation is random, but the totals are not affected by an imputation variance. [Copyright &y& Elsevier]

The cube method allows the selection of balanced samples on several auxiliary variables with equal or unequal inclusion probabilities. Practical implementation of the cube method has raised questions concerning the selection of a multi-phase balanced sampling design, the rebalancing of an unbalanced sampling design by completing it with another sample, the selection of a balanced sample from an unbalanced sample and the coordination of balanced samples. This paper provides a complete solution of all these problems. [ABSTRACT FROM AUTHOR]

The Neyman–Scott rectangular pulses model is a clustered point process in time. This article presents a new method of parameters estimation for this model applied to rainfall simulation. It is based on a modified method of moments using two temporal scales of aggregation. A simple algebraic computation leads to reduce the number of parameters estimated by minimisation. Indeed two parameters are obtained by minimisation while the three others can be directly computed. The optimisation method used is based on the Nelder–Mead simplex. The minimisation procedure is stable with regard to the starting point and always converges. The related approximate confidence interval is obtained using the delta method and block bootstrap techniques. The validation is carried out using the rainfall station of Payerne situated on the Swiss Plateau. 100series of 10 years of hourly rainfall have been generated for both a summer and a winter month. From the obtained simulated series 100 sets of parameters have been determined. These parameters estimators are unbiased. [Copyright &y& Elsevier]

High densities of local-scale daily precipitation series across relatively large domains are of special interest for a wide range of applications (e.g., hydrological modeling, agriculture). The focus of the present study is to post-process gridded precipitation from a single reanalysis to correct bias and scale mismatch with observations, and to extend the same post-processing at sites without historical data. A Stochastic Model Output Statistical approach combined with meta-Gaussian spatiotemporal random fields, calibrated at sites, is employed to post-process the Climate Forecast System Reanalysis (CFSR) precipitation. The post-processed data, characterized by local parameters, is then mapped across the Great Lakes region (Canada) using two different approaches: (1) kriging, and (2) Vector Generalized Additive Model (VGAM) with spatial covariates. The kriging enables the interpolation of these parameters, while the spatial VGAM helps to spatially post-process CFSR precipitation using a single model. The k-fold cross-validation procedure is employed to assess the ability of the two approaches to predict selected characteristics and climate indices. The kriging and spatial VGAM approaches modeled effectively the distribution of the precipitation process to similar extents (e.g., mean daily precipitation, variability and the number of wet days). The kriging approach produces slightly better estimates of climate indices than the spatial VGAM models. Both approaches demonstrate significant improvement of the metric estimation compared to those of CFSR without post-processing. [ABSTRACT FROM AUTHOR]

Reanalyses, generated by numerical weather prediction methods assimilating past observations, provide consistent and continuous meteorological fields for a specific period. In regard to precipitation, reanalyses cannot be used as a climate proxy of the observed precipitation, as biases and scale mismatches exist between the datasets. In the present study, a stochastic model output statistics (SMOS) approach combined with meta-Gaussian spatiotemporal random fields was employed to cope with these caveats. The SMOS is based on the generalized linear model (GLM) and the vector generalized linear model (VGLM) frameworks to model the precipitation occurrence and intensity, respectively. Both models use the Climate Forecast System Reanalysis (CFSR) precipitation as covariate and were locally calibrated at 173 sites across the Great Lakes region. Combined with meta-Gaussian random fields, the GLM and VGLM models allowed for the generation of spatially coherent daily precipitation fields across the region. The results indicated that the approach corrected systematic biases and provided an accurate spatiotemporal structure of daily precipitation. Performances of selected precipitation indicators from the joint Commission for Climatology (CCl)/CLIVAR/JCOMM Expert Team on Climate Change Detection and Indices (ETCCDI) were good and were systematically improved when compared to CFSR. [ABSTRACT FROM AUTHOR]

This paper develops a new nonparametric method that is suitable for detecting slowly-varying nonstationarities that can be seen as trends in the time marginal of the time-varying spectrum of the signal. The rationale behind the proposed method is to measure the importance of the trend in the time marginal by using a proper test statistic, and further compare this measurement with the ones that are likely to be found in stationary references. It is shown that the distribution of the test statistic under stationarity can be modeled fairly well by a Generalized Extreme Value (GEV) pdf, from which a threshold can be derived for testing stationarity by means of a hypothesis test. Finally, the new method is compared with other ones found in the literature. [ABSTRACT FROM AUTHOR]

Traditional design flood estimation approaches have focused on peak discharges and have often neglected other hydrograph characteristics such as hydrograph volume and shape. Synthetic design hydrograph estimation procedures overcome this deficiency by jointly considering peak discharge, hydrograph volume, and shape. Such procedures have recently been extended to allow for the consideration of process variability within a catchment by a flood-type specific construction of design hydrographs. However, they depend on observed runoff time series and are not directly applicable in ungauged catchments where such series are not available. To obtain reliable flood estimates, there is a need for an approach that allows for the consideration of process variability in the construction of synthetic design hydrographs in ungauged catchments. In this study, we therefore propose an approach that combines a bivariate index flood approach with event-type specific synthetic design hydrograph construction. First, regions of similar flood reactivity are delineated and a classification rule that enables the assignment of ungauged catchments to one of these reactivity regions is established. Second, event-type specific synthetic design hydrographs are constructed using the pooled data divided by event type from the corresponding reactivity region in a bivariate index flood procedure. The approach was tested and validated on a dataset of 163 Swiss catchments. The results indicated that 1) random forest is a suitable classification model for the assignment of an ungauged catchment to one of the reactivity regions, 2) the combination of a bivariate index flood approach and event-type specific synthetic design hydrograph construction enables the consideration of event types in ungauged catchments, and 3) the use of probabilistic class memberships in regional synthetic design hydrograph construction helps to alleviate the problem of misclassification. Event-type specific synthetic design hydrograph sets enable the inclusion of process variability into design flood estimation and can be used as a compromise between single best estimate synthetic design hydrographs and continuous simulation studies. [ABSTRACT FROM AUTHOR]

Estimates of flood event magnitudes with a certain return period are required for the design of hydraulic structures. While the return period is clearly defined in a univariate context, its definition is more challenging when the problem at hand requires considering the dependence between two or more variables in a multivariate framework. Several ways of defining a multivariate return period have been proposed in the literature, which all rely on different probability concepts. Definitions use the conditional probability, the joint probability, or can be based on the Kendall's distribution or survival function. In this study, we give a comprehensive overview on the tools that are available to define a return period in a multivariate context. We especially address engineers, practitioners, and people who are new to the topic and provide them with an accessible introduction to the topic. We outline the theoretical background that is needed when one is in a multivariate setting and present the reader with different definitions for a bivariate return period. Here, we focus on flood events and the different probability concepts are explained with a pedagogical, illustrative example of a flood event characterized by the two variables peak discharge and flood volume. The choice of the return period has an important effect on the magnitude of the design variable quantiles, which is illustrated with a case study in Switzerland. However, this choice is not arbitrary and depends on the problem at hand. WIREs Water 2016, 3:819-833. doi: 10.1002/wat2.1173 For further resources related to this article, please visit the . [ABSTRACT FROM AUTHOR]

Summary A rating curve is used to indirectly estimate the discharge in rivers based on water level measurements. The discharge values obtained from a rating curve include uncertainties related to the direct stage–discharge measurements (gaugings) used to build the curves, the quality of fit of the curve to these measurements and the constant changes in the river bed morphology. Moreover, the uncertainty of discharges estimated from a rating curve increases with the “age” of the rating curve. The level of uncertainty at a given point in time is therefore particularly difficult to assess. A “dynamic” method has been developed to compute rating curves while calculating associated uncertainties, thus making it possible to regenerate streamflow data with uncertainty estimates. The method is based on historical gaugings at hydrometric stations. A rating curve is computed for each gauging and a model of the uncertainty is fitted for each of them. The model of uncertainty takes into account the uncertainties in the measurement of the water level, the quality of fit of the curve, the uncertainty of gaugings and the increase of the uncertainty of discharge estimates with the age of the rating curve computed with a variographic analysis ( Jalbert et al., 2011 ). The presented dynamic method can answer important questions in the field of hydrometry such as “How many gaugings a year are required to produce streamflow data with an average uncertainty of X%?” and “When and in what range of water flow rates should these gaugings be carried out?”. The Rocherousse hydrometric station (France, Haute-Durance watershed, 946 [km 2 ]) is used as an example throughout the paper. Others stations are used to illustrate certain points. [ABSTRACT FROM AUTHOR]

Statistical procedures for the detection of a change in the dependence structure of a series of multivariate observations are studied in this work. The test statistics that are proposed are $L_1$, $L_2$, and $L_{\infty }$ distances computed from vectors of differences of Kendall's tau; two multivariate extensions of Kendall's measure of association are used. Since the distributions of these statistics under the null hypothesis of no change depend on the unknown underlying copula of the vectors, a procedure based on the multiplier central limit theorem is used for the computation of p-values; the method is shown to be valid both asymptotically and for moderate sample sizes. Alternative versions of the tests that take into account possible breakpoints in the marginal distributions are also investigated. Monte Carlo simulations show that the tests are powerful under many scenarios of change-point. In addition, two estimators of the time of change are proposed and their efficiency is carefully studied. The methodologies are illustrated on simulated series from the Canadian Regional Climate Model. The Canadian Journal of Statistics 41: 65-82; 2013 © 2012 Statistical Society of Canada [ABSTRACT FROM AUTHOR]

Summary: A rating curve provides an estimation of river discharges based on stage (water level). This estimation contains a level of uncertainty. Initial uncertainty occurs at the time of establishment of the rating curve. This may be due, for example, to the randomness of natural processes or to the inaccurate measurement of the stage. Temporal uncertainty is related to the well-known processes of erosion and deposition that modify the geometry of the river bed and, consequently, the relationship between the stage and discharge. As time goes by, temporal uncertainty of the estimated discharge from a rating curve increases. Due to the widespread use of rating curves by scientists and water resource managers, it is important to assess these related uncertainties. Several studies have taken into account initial uncertainties but none, to our knowledge, has considered temporal uncertainties. The aim of this paper is to develop a methodology to estimate the temporal uncertainty of the discharge that is estimated by the rating curve. The proposed approach is based on a variographic analysis. At the beginning of rating curve validity period, the estimated discharge is believed to be distributed as a normal distribution centered on the rating curve’s estimation. The initial variance of the normal distribution, according to the initial uncertainty, is fixed so that the relative uncertainty is less than 5%. A temporal variance term, estimated using a variographic analysis, is then added to the initial variance to take into account the temporal uncertainty. This term corresponds to the mean of semi-variance between estimations separated by a given time. The proposed method has been applied to 1803 gaugings from 19 hydrometric stations located in the French Alps. The 95% confidence intervals cover 90% of 1803 gaugings. This result shows that the confidence intervals are too short. However, this may be due to an underestimation of the initial variance. The method is efficient and robust since it can be adapted to various station characteristics, such as trends in discharge series or stability of the river bed. [Copyright &y& Elsevier]

In Amazonia, because the small catchments are ungauged, it is not possible to analyse them, for example, for hydroelectric power production. Thus, the objective of this paper is to study the transferability of a rainfall-runoff model to simulate flow duration curves for the production of hydroelectric power. The approach is based on the transfer of the impulse response of a model calibrated on two gauged catchments, allowing the evaluation approach permutation between these two catchments. We have, respectively, 7 years and 2 years and 2 months of rainfall and runoff data for these catchments. A sensitivity analysis of the transferability calibration to the sample size is carried out to determine the shortest flow period gauged on the receptor catchment, which produces results comparable to those calibrated with the maximum samples size. This analysis evaluates fieldwork on the ungauged sites of the region. En Amazonie, les petits bassins versants ne disposent pas de données de débit, ce qui empêche l’étude de ces bassins, par exemple, pour des fins de production d’énergie hydroélectrique. Ainsi, l’objectif de ce travail est d’étudier la transférabilité d’un modèle pluie-débit pour simuler les débits classés en vue d’une planification de la production hydroélectrique. L’approche est basée sur le transfert de la réponse impulsionnelle d’un modèle étalonné sur deux bassins jaugés permettant l’évaluation de la permutation de l’approche. Sur ces bassins test, on dispose respectivement de 7 ans et de 2 ans et 2 mois de données de pluie et de débit. L’analyse de sensibilité de l’étalonnage de la transférabilité détermine la taille minimale de la série de débits mesurés sur le bassin cible qui assure encore des résultats comparables à ceux étalonnés avec les échantillons de taille maximale de débits. Celle-ci évalue l’effort de mesures sur des sites non jaugés de la région. [ABSTRACT FROM AUTHOR]

In Amazonia, hydroelectric power production has only been developed on large basins, which are the only ones that have been gauged. This excludes innumerable small catchments, for which only rainfall data are available. Therefore, the objective of the work presented in this paper is to develop a hydrological rainfall–runoff model to simulate flow duration curves for hydro power production planning. The model is based on a linear and time-invariant system (input–output). The impulse response of the system is calculated from the cross-spectral analysis between the rainfall and runoff series. This response is optimized successive approximations to minimize the root mean square of the error. The test catchment area has 7 years of rainfall and runoff data; 4 years are considered for the calibration and 3 years for the validation of the model. A sensitivity analysis of the model to the sample size is carried out to determine the shortest possible data period that produces results comparable to those of the model validation. [ABSTRACT FROM AUTHOR]

Design flood estimates for a given return period are required in both gauged and ungauged catchments for hydraulic design and risk assessments. Contrary to classical design estimates, synthetic design hydrographs provide not only information on the peak magnitude of events but also on the corresponding hydrograph volumes together with the hydrograph shapes. In this study, we tested different regionalization approaches to transfer parameters of synthetic design hydrographs from gauged to ungauged catchments. These approaches include classical regionalization methods such as linear regression techniques, spatial methods, and methods based on the formation of homogeneous regions. In addition to these classical approaches, we tested nonlinear regression models not commonly used in hydrological regionalization studies, such as random forest, bagging, and boosting. We found that parameters related to the magnitude of the design event can be regionalized well using both linear and nonlinear regression techniques using catchment area, length of the main channel, maximum precipitation intensity, and relief energy as explanatory variables. The hydrograph shape, however, was found to be more difficult to regionalize due to its high variability within a catchment. Such variability might be better represented by looking at flood-type specific synthetic design hydrographs. [ABSTRACT FROM AUTHOR]

Summary Streamflow time series provide baseline data for many hydrological investigations. Errors in the data mainly occur through uncertainty in gauging (measurement uncertainty) and uncertainty in the determination of the stage-discharge relationship based on gaugings (rating curve uncertainty). As the velocity-area method is the measurement technique typically used for gaugings, it is fundamental to estimate its level of uncertainty. Different methods are available in the literature ( ISO 748 , Q + , IVE ), all with their own limitations and drawbacks. Among the terms forming the combined relative uncertainty in measured discharge, the uncertainty component relating to the limited number of verticals often includes a large part of the relative uncertainty. It should therefore be estimated carefully. In ISO 748 standard, proposed values of this uncertainty component only depend on the number of verticals without considering their distribution with respect to the depth and velocity cross-sectional profiles. The Q + method is sensitive to a user-defined parameter while it is questionable whether the IVE method is applicable to stream-gaugings performed with a limited number of verticals. To address the limitations of existing methods, this paper presents a new methodology, called FLow Analog UnceRtainty Estimation (F laure ), to estimate the uncertainty component relating to the limited number of verticals. High-resolution reference gaugings (with 31 and more verticals) are used to assess the uncertainty component through a statistical analysis. Instead of subsampling purely randomly the verticals of these reference stream-gaugings, a subsampling method is developed in a way that mimicks the behavior of a hydrometric technician. A sampling quality index (SQI) is suggested and appears to be a more explanatory variable than the number of verticals. This index takes into account the spacing between verticals and the variation of unit flow between two verticals. To compute the uncertainty component for any routine gauging, the four most similar gaugings among the reference stream-gaugings dataset are selected using an analog approach, where analogy includes both riverbed shape and flow distribution complexity. This new method was applied to 3185 stream-gaugings with various flow conditions and compared with the other methods ( ISO 748 , IVE , Q + with a simple automated parametrization). Results show that F laure is overall consistent with the Q + method but not with ISO 748 and IVE methods, which produce clearly overestimated uncertainties for discharge measurements with less than 15 verticals. The F laure approach therefore appears to be a consistent method. An advantage is the explicit link made between the estimation of cross-sectional interpolation errors and the study of high-resolution reference gaugings. [ABSTRACT FROM AUTHOR]

A correction to the article regarding the name of the third coauthor that was published in the 2014 issue is presented.

Summary: The Peak-Over-Threshold (POT) approach is an interesting alternative to the one based on Annual Maxima (AM) series since it gives the opportunity to take into consideration extreme events that would not be considered otherwise. It has also been recognized that the regional approach improves statistical inference when compared to the local approach, assuming that the region is statistically homogeneous. A regional POT approach was developed and applied to the network stations located in southern Québec. POT series for 5-, 10-, 15-, 30-min and 1-, 2-, 6- and 12-h durations were constructed assuming a fixed exceedance rate. An analysis of local POT series showed that the intra-annual variability of the Generalized Pareto Distribution (GPD) parameters needs to be taken into consideration. Models of various complexities were defined combining local and regional representations as well as the intra-annual variability of GPD parameters. Regional likelihood was estimated and models were compared based on the Akaike Information Criterion (AIC). Models with regional shape and scale parameters and accounting for intra-annual variability were selected for all durations. Spatial covariates were also introduced through a simple model linking GPD parameters to latitude, longitude and altitude. The sensitivity of results to threshold values and selected models was also investigated. Interpolated maps of intense rainfall over the studied area are finally proposed. [ABSTRACT FROM AUTHOR]

The safe and cost-effective design of flood protection structures requires reliable flood estimates. Such estimates usually describe a flood in terms of peak discharge but neglect other flood characteristics such as flood volume and the interdependence between these characteristics. Furthermore, non-stationarities in hydrological time series and in particular the dependence between peak discharge and flood volume due to climate change are often neglected. This study quantifies trends in peak discharge, flood volume, and the dependence between the two variables on a spatially comprehensive dataset of 307 medium-sized catchments in Switzerland. In addition, it identifies potential climatological drivers for changes in these two variables and their dependence. To do so, the hydrological model PREVAH is driven with future meteorological time series given by the Swiss EURO-CORDEX based CH2018 climate scenarios. Flood and climatological event characteristics are extracted from the resulting discharge time series and analyzed for trends. The results show spatially heterogeneous trends in both the dependence between peak discharge and flood volume as well as the potential climatological drivers precipitation, snowmelt, and soil moisture. Trends in the climatological drivers can partly explain trends in the dependence between peak discharge and flood volume only if they are jointly considered. The rather positive trend in dependence in the Alps can partly be explained by a decrease in snowmelt and an increase in precipitation. The negative trend in dependence in the Plateau region can in some catchments partly also be explained by changes in snowmelt while it can be explained by changes in soil moisture in other catchments. Trends in dependence but also the links between drivers and trends are uncertain since they depend on the climate chain considered. It remains to be discussed whether such uncertain, expected changes in the dependence between peak discharge and flood volume should be considered in modeling efforts. [ABSTRACT FROM AUTHOR]