Charles Obled

The sensitivity of hydrological models to spatial rainfall patterns: an evaluation using observed data

Spatial variability of rainfall is often considered as a major source of temporal variability in the resulting basin hydrograph. Since direct experimental evidence is not available, this must be verified through a modelling approach, provided adequate data are available. A semidistributed version of TOPMODEL has therefore been applied to the Real Collobrier experimental basin (71 km2 in southeast France with 21 recording raingauges) using an hourly time step and a series of independent events. First, a set of reference results has been built under the assumption of spatial uniformity for the rainfall. Two different densities of network have been tested (including 5 or 21 gauges), showing a significant advantage for the dense network rainfall estimate. Next, the spatial variability of the rainfall field has been tested and confirmed, with commonly a factor of 3 between simultaneous average rainfall over subcatchment areas of 6–8 km2. However, the model response reflects this spatial variability only in secondary peaks which are usually an order of magnitude smaller than the bulk of the hydrograph and not always present in the observed discharges. An extended discussion considers if these results may be dependent on the model or on the setting up of the numerical experiments. In fact, it seems that the spatial variability of rainfall, although important, is not sufficiently organized in time and space to overcome the effects of smoothing and dampening when running off through this rural medium-sized catchment. Such results may not hold for smaller urbanized areas or larger rural basins.

Physical interpretation and sensitivity analysis of the TOPMODEL

The TOPMODEL is a variable contributing area conceptual model in which the predominant factors determining the formation of runoff are represented by the topography of the basin and a negative exponential law linking the transmissivity of the soil with the distance to the saturated zone below the ground level. Although conceptual, this model is frequently described as a ‘physically based model’ in the sense that its parameters can be measured directly in situ. In line with the analysis of various conceptual rainfall-runoff models conducted by Franchini and Pacciani (J. Hydrol., 122: 161–219, 1991), a detailed analysis of the TOPMODEL is performed to arrive at a closer understanding of the correspondence of the assumptions underpinning the model with the physical reality and, in particular, the role that topographic information (expressed by the topographic index curve) and the nature of the soil (expressed by saturated hydraulic conductivity and its decay with soil depth), have within the model itself. Also investigated is the extent to which the model parameters actually reflect the physical properties to which they refer and how far their values offset the inevitable schematisation of the model. The various applications to real situations include the Sieve basin (river Arno tributary), which was used for the comparison of conceptual rainfall-runoff models described in the above-mentioned study by Franchini and Paccini. This allows that analysis to be extended to the TOPMODEL.

Quantitative precipitation forecasts: a statistical adaptation of model outputs through an analogues sorting approach

Abstract Medium-term quantitative precipitation forecasts (QPFs) up to several days ahead are required to issue early flood warnings and to allow optimum operation of hydraulic structures or reservoirs. This paper describes an approach which can be seen as an adaptation of deterministic meteorological model outputs. It involves searching for a sample of past situations similar to the current one from a long meteorological archive. The analogy is considered in terms of general circulation patterns over a window covering western Europe. For this restricted sample of days similar to the day at hand, the corresponding sample of observed daily precipitation is extracted for each catchment. The rainfall to be observed during the current day is assumed to follow the same distribution, known from this empirical sample. This provides a probabilistic forecast expressed, for example, by a central quantile and a confidence range. This paper describes the many choices underlying the optimisation of this approach: choice of predictor variables to characterise a meteorological situation, choice of similarity criterion between two situations, criterion for performance evaluation between two versions of the algorithm, etc. This method was calibrated over about 50 catchments located in France, Italy and Spain, using a meteorological and hydrological archive running from 1953 to 1996. Comparisons carried out over a validation sample (1995–1996) with three poor-man methods prove the interest of this approach, in a perfect prognosis context. In real-time operation, the use of forecast instead of observed predictor variables, essentially geopotential fields, produces only a minor decrease in performance. The use of the single-valued central quantile supplemented by the confidence interval provided a QPF that has proved effective and informative on the potential for extreme values.

Including spatially variable effective soil depths in TOPMODEL

Abstract TOPMODEL ( Beven and Kirkby, 1979 ; Beven et al., 1995 ) was one of the first attempts to model distributed hydrological responses based on variable contributing area concepts. It makes use of an index of hydrological similarity based on an analysis of the topographic data. The index approach was later generalised to take account of spatial variability of soil transmissivities, but no similar spatial analysis of the variability in the rate of the decrease of the transmissivity with depth has yet been examined. This paper shows how the TOPMODEL theory can be extended to handle this spatial variability, using a 2D distribution function of a new soil depth-topographic index of hydrological similarity. A first sensitivity analysis of the effect of variable soil depths on the model predictions for the Maurets catchment, France, is presented. Predicted discharges and calibrated parameter values are not sensitive to the patterns of effective soil depth investigated. Distributed predictions may be more sensitive but raise questions of how to obtain the parameter data required.

Analytical compensation between DTM grid resolution and effective values of staurated hydraulic conductivity within the TOPMODEL framework

The recent widespread availability of digital terrain data has resulted in increasing use, in a variety of hydrological models, of different degrees of complexity. Previous experience suggests that neither model results nor effective parameter values are independent of the resolution of the digital terrain data used. Calibration of parameter values can compensate for lack of resolution in the digital terrain data. This paper will show, for one particular model, TOPMODEL, how an analytical link can be established between the grid size of a raster digital terrain model and the effective saturated hydraulic conductivity value used in the model. The work generalizes the results of the recent study by Franchini et al. (1996) and allows the change in effective conductivity to be estimated on the basis of keeping a realistic simulation of saturated contributing areas as the DTM grid size changes.

Digital elevation analysis for distributed hydrological modeling: Reducing scale dependence in effective hydraulic conductivity values

The recent widespread availability of digital terrain data has made automatic procedures for topographic analyses popular. Previous studies have shown that hydrological models and their effective parameter values are dependent on the resolution of the elevation grid. This paper examines the analysis of raster elevation data within the topography-based model, TOPMODEL, framework. It is demonstrated that the algorithm used in processing channel pixels in calculating the topographic index k = ln(a/tanβ) can have a dramatic effect on the sensitivity of effective parameter values to the grid size. Suggestions are made for calculating the topographic index of channel pixels, consistent with the TOPMODEL assumptions, that strongly decrease the sensitivity of the calibrated effective hydraulic conductivity values to grid size.

Toward Real-Time Daily PQPF by an Analog Sorting Approach: Application to Flash-Flood Catchments

Heavy-rainfall events are common in southern France and frequently result in devastating flash floods. Thus, an appropriate anticipation of future rainfall is required: for early flood warning, at least 12‐24 h in advance; for alerting operational services, at least 2‐3 days ahead. Precipitation forecasts are generally provided by numerical weather prediction models (NWP), and their associated uncertainty is generally estimatedthroughan ensemble approach. Precipitation forecastsalso have tobe adaptedto hydrologicalscales. This study describes an alternative approach to commonly used limited-area models. Probabilistic quantitative precipitation forecasts (PQPFs) are provided through an analog sorting technique, which directly links synoptic-scale NWPoutput to catchment-scale rainfallprobabilitydistributions. One issue concerns the latest developments in implementing a daily version of this technique into operational conditions. It is shown that the obtained PQPFs depend on the meteorological forecasts used for selecting analogous days and that the method has to be reoptimized when changing the source of synoptic forecasts, because of the NWP output uncertainties. Second, an evaluation of the PQPFs demonstrates that the analog technique performs well for early warning of heavy-rainfall events and provides useful information as potential input to a hydrological ensemble prediction system. It is shown that the obtained daily rainfall distributions can be unreliable. A statistical correction of the observed bias is proposed as a function of the no-rain frequency values, leading to a significant improvement in PQPF sharpness.

Spatial Similarity and Transferability of Analog Dates for Precipitation Downscaling over France

AbstractHigh-resolution weather scenarios generated for climate change impact studies from the output of climate models must be spatially consistent. Analog models (AMs) offer a high potential for the generation of such scenarios. For each prediction day, the scenario they provide is the weather observed for days in a historical archive that are analogous according to different predictors. When the same “analog date” is chosen for a prediction at several sites, spatial consistency is automatically satisfied. The optimal predictors and consequently the optimal analog dates, however, are expected to depend on the location for which the prediction is to be made.In the present work, the predictor (1000- and 500-hPa geopotential heights) domain of a benchmark AM is optimized for the probabilistic daily prediction of 8981 local precipitation “stations” over France. The corresponding 8981 locally domain-optimized AMs are used to explore the spatial transferability and similarity of the optimal analog dates obtai...

VERS UN SYSTÈME OPÉRATIONNEL DE PRÉVISION NUMÉRIQUE DES AVALANCHES [Agrave] PARTIR DE MÉTHODES STATISTIQUES

ABSTRACT This paper is a review of the research work being undertaken in the ‘Institut National Polytechnique de Grenoble’ in close cooperation with ‘Institut fur Schnee und Lawinenforschung’ of Davos (Switzerland). Although the authors are primarily concerned with hydrological forecasting, it appeared that some of the methods employed could be transposed and adapted to avalanche forecasting. The avalanche phenomena, whose physics is not yet fully understood can be roughly described as follows (at least for forecasting purposes): a conditioning of the snowpack by meteorological variables (snowfall, wind, temperature …) at a regional scale. a local triggering of the snowlayer, caused by mechanical or thermal effects and much more randomly distributed. The goal of the research was to derive some relationships between the parameters previously described and the occurrences of avalanches for the purpose of their better understanding and forecasting.