scholarly journals On the potential of variational calibration for a fully distributed hydrological model: application on a Mediterranean catchment

2019 ◽  
Author(s):  
Maxime Jay-Allemand ◽  
Pierre Javelle ◽  
Igor Gejadze ◽  
Patrick Arnaud ◽  
Pierre-Olivier Malaterre ◽  
...  
Keyword(s):  
Hydrological Model ◽  
Weather Forecasting ◽  
Flash Flood ◽  
Estimation Method ◽  
Distributed Hydrological Model ◽  
Initial State ◽  
Ungauged Catchments ◽  
Temporal Dimension ◽  
Radar Rainfall ◽  
Additional Constraints

Abstract. Flash flood alerts in metropolitan France are provided by SCHAPI (Service Central Hydrométéorologique et d’Appui à la Prévision des Inondations) through the Vigicrues Flash service, which is designed to work in ungauged catchments. The AIGA method implemented in Vigicrues Flash is designed for flood forecasting on small- and medium-scale watersheds. It is based on a distributed hydrological model accounting for spatial variability of the rainfall and the catchment properties, based on the radar rainfall observation inputs. Calibration of distributed parameters describing these properties with high resolution is difficult, both technically (in terms of the estimation method), and because of the identifiability issues. Indeed, the number of parameters to be calibrated is much greater than the number of spatial locations where the discharge observations are usually available. However, the flood propagation is a dynamic process, so observations have also a temporal dimension. This must be larger enough to comprise a representative set of events. In order to fully benefit from using the AIGA method, we consider its hydrological model (GRD) in combination with the variational estimation (data assimilation) method. In this method, the optimal set of parameters is found by minimizing the objective function which includes the misfit between the observed and predicted values and some additional constraints. The minimization process requires the gradient of the cost function with respect to all control parameters, which is efficiently computed using the adjoint model. The variational estimation method is scalable, fast converging, and offers a convenient framework for introducing additional constraints relevant to hydrology. It can be used both for calibrating the parameters and estimating the initial state of the hydrological system for short range forecasting (in a manner used in weather forecasting). The study area is the Gardon d’Anduze watershed where four gauging stations are available. In numerical experiments, the benefits of using the distributed against the uniform calibration are analysed in terms of the model predictive performance. Distributed calibration shows encouraging results with better model prediction at gauged and ungauged locations.

scholarly journals Flash Flood Simulation for Ungauged Catchments Based on the Distributed Hydrological Model

Water ◽  
2019 ◽  
Vol 11 (1) ◽  
pp. 76 ◽  
Author(s):  
Pengfei Jia ◽  
Ronghua Liu ◽  
Meihong Ma ◽  
Qi Liu ◽  
Yali Wang ◽  
...  
Keyword(s):  
Early Warning ◽  
Hydrological Model ◽  
Flash Flood ◽  
Model Simulation ◽  
Model Parameters ◽  
Distributed Hydrological Model ◽  
Ungauged Catchments ◽  
Runoff Water ◽  
Effective Measure ◽  
Warning Effect

Flash flood is a significant threat to those who live in China and beyond. Reproducing them with distributed hydrological models is an effective measure for preventing flash floods. This paper introduces the China flash flood hydrological model (CNFF-HM); explores its principles, key steps, and applicability; and validates its simulation effects in two typical, small ungauged catchments (UCs), namely, Tiezhuling and Qianyangxi. First, the applicability of the model and the portability of the parameters were verified in Qianyangxi. Then, in Tiezhuling, the model parameters were comprehensively determined based on the measured values and simulation results, and the influence of the reservoir on the simulation effect was analyzed. Using the model’s different parameters in different catchments, the characteristics of the basin were compared and analyzed from the aspects of evapotranspiration, runoff, water distribution, and confluence. Finally, the early warning effect was evaluated. The results indicated the following: (1) The reservoir has a significant impact on model simulation. (2) The two UCs have similarities in evapotranspiration, runoff, and water source, and the difference in convergence is obvious. (3) With better early warning effect, the warning can achieve full coverage. These positive results suggest that this method should be further developed and tested.

scholarly journals Methodology for post-event analysis of flash floods - Svacenický Creek case study

2011 ◽  
Vol 41 (3) ◽  
pp. 235-250 ◽  
Author(s):  
Lotta Blaškovičová ◽  
Oliver Horvát ◽  
Kamila Hlavčová ◽  
Silvia Kohnová ◽  
Ján Szolgay
Keyword(s):  
Hydrological Model ◽  
Flash Flood ◽  
Curve Number ◽  
Flash Floods ◽  
Event Analysis ◽  
Distributed Hydrological Model ◽  
Radar Rainfall ◽  
Flood Peak ◽  
Spatially Distributed

Methodology for post-event analysis of flash floods - Svacenický Creek case study In this paper a methodology for a post-event analysis of a flash flood and estimation of the flood peak and volume are developed and tested. The selected flash flood occurred on the 6th of June, 2009 in the Svacenický Creek Basin. To understand rainfall-runoff processes during this extreme flash flood, the runoff response was simulated using the spatially-distributed hydrological model KLEM (Kinematic Local Excess Model). The distributed hydrological model is based on the availability of raster information about the landscape's topography, soil and vegetation properties and radar rainfall data. In the model, the SCS-Curve Number procedure is applied to a grid for the spatially-distributed representation of the runoff-generating processes. A description of the drainage system's response is used to represent the runoff's routing. The simulated values achieved by the KLEM model were comparable with the maximum peak estimated on the basis of the post-event surveying. The consistency of the estimated and simulated values from the KLEM model was evident both in time and space, and the methodology has shown its practical applicability.

scholarly journals Calibration of physically based models: back to basics?

2003 ◽  
Vol 5 (4) ◽  
pp. 233-244 ◽  
Author(s):  
Vincent Guinot ◽  
Philippe Gourbesville
Keyword(s):  
Hydrological Model ◽  
Flash Flood ◽  
Model Parameters ◽  
Model Structure ◽  
Ungauged Catchments ◽  
Model Response ◽  
Physically Based ◽  
Physically Based Models ◽  
Extreme Hydrological Events ◽  
Main Factor

The modelling of extreme hydrological events often suffers from a lack of available data. Physically based models are the best available modelling option in such situations, as they can in principle provide answers about the behaviour of ungauged catchments provided that the geometry and the forcings are known with sufficient accuracy. The need for calibration is therefore limited. In some situations, calibration (seen as adjusting the model parameters so that they fit the calculation as closely to the measurements as possible) is impossible. This paper presents such a situation. The MIKE SHE physically based hydrological model is used to model a flash flood over a medium-sized catchment of the Mediterranean Alps (2820 km2). An examination of a number of modelling alternatives shows that the main factor of uncertainty in the model response is the model structure (what are the dominant processes). The second most important factor is the accuracy with which the catchment geometry is represented in the model. The model results exhibit very little sensitivity to the model parameters, and therefore calibration of these parameters is found to be useless.

scholarly journals Application of a distributed hydrological model to the design of a road inundation warning system for flash flood prone areas

2010 ◽  
Vol 10 (4) ◽  
pp. 805-817 ◽  
Author(s):  
P.-A. Versini ◽  
E. Gaume ◽  
H. Andrieu
Keyword(s):  
Cross Sections ◽  
Hydrological Model ◽  
Flash Flood ◽  
Warning System ◽  
Flash Floods ◽  
Probability Of Detection ◽  
Distributed Hydrological Model ◽  
Rainfall Runoff ◽  
Erosion Processes ◽  
Stage Of Development

Abstract. This paper presents an initial prototype of a distributed hydrological model used to map possible road inundations in a region frequently exposed to severe flash floods: the Gard region (South of France). The prototype has been tested in a pseudo real-time mode on five recent flash flood events for which actual road inundations have been inventoried. The results are promising: close to 100% probability of detection of actual inundations, inundations detected before they were reported by the road management field teams with a false alarm ratios not exceeding 30%. This specific case study differs from the standard applications of rainfall-runoff models to produce flood forecasts, focussed on a single or a limited number of gauged river cross sections. It illustrates that, despite their lack of accuracy, hydro-meteorological forecasts based on rainfall-runoff models, especially distributed models, contain valuable information for flood event management. The possible consequences of landslides, debris flows and local erosion processes, sometimes associated with flash floods, were not considered at this stage of development of the prototype. They are limited in the Gard region but should be taken into account in future developments of the approach to implement it efficiently in other areas more exposed to these phenomena such as the Alpine area.

scholarly journals Post-event analysis and flash flood hydrology in Slovakia

2016 ◽  
Vol 64 (4) ◽  
pp. 304-315 ◽  
Author(s):  
Kamila Hlavčová ◽  
Silvia Kohnová ◽  
Marco Borga ◽  
Oliver Horvát ◽  
Pavel Šťastný ◽  
...  
Keyword(s):  
Hydrological Model ◽  
Flash Flood ◽  
Flash Floods ◽  
Event Analysis ◽  
Distributed Hydrological Model ◽  
Rainfall Runoff ◽  
Flood Events ◽  
Spatially Distributed ◽  
Catchment Size ◽  
Flood Peaks

Abstract This work examines the main features of the flash flood regime in Central Europe as revealed by an analysis of flash floods that have occurred in Slovakia. The work is organized into the following two parts: The first part focuses on estimating the rainfall-runoff relationships for 3 major flash flood events, which were among the most severe events since 1998 and caused a loss of lives and a large amount of damage. The selected flash floods occurred on the 20th of July, 1998, in the Malá Svinka and Dubovický Creek basins; the 24th of July, 2001, at Štrbský Creek; and the 19th of June, 2004, at Turniansky Creek. The analysis aims to assess the flash flood peaks and rainfall-runoff properties by combining post-flood surveys and the application of hydrological and hydraulic post-event analyses. Next, a spatially-distributed hydrological model based on the availability of the raster information of the landscape’s topography, soil and vegetation properties, and rainfall data was used to simulate the runoff. The results from the application of the distributed hydrological model were used to analyse the consistency of the surveyed peak discharges with respect to the estimated rainfall properties and drainage basins. In the second part these data were combined with observations from flash flood events which were observed during the last 100 years and are focused on an analysis of the relationship between the flood peaks and the catchment area. The envelope curve was shown to exhibit a more pronounced decrease with the catchment size with respect to other flash flood relationships found in the Mediterranean region. The differences between the two relationships mainly reflect changes in the coverage of the storm sizes and hydrological characteristics between the two regions.

scholarly journals Application of deterministic distributed hydrological model for large catchment: a case study at Vu Gia Thu Bon catchment, Vietnam

2016 ◽  
Vol 18 (5) ◽  
pp. 885-904 ◽  
Author(s):  
Ngoc Duong Vo ◽  
Philippe Gourbesville
Keyword(s):  
Overland Flow ◽  
Hydrological Model ◽  
Unsaturated Flow ◽  
Correlation Coefficients ◽  
Water Levels ◽  
Distributed Hydrological Model ◽  
Ungauged Catchments ◽  
Quality Of Results ◽  
The Impact

In order to create a tool to help hydrologists and authorities to have good understanding about occurrences in stream flow regime together with its variation in the future under the impact of climate change in the Vu Gia Thu Bon catchment, a deterministic distributed hydrological model has been developed and constructed. This model covers the major processes in the hydrologic cycle including rainfall, evapotranspiration, overland flow, unsaturated flow, groundwater flow, channel flow, and their interactions. The model is calibrated and validated against the daily data recorded at seven stations during 1991–2000 and 2001–2010, respectively. The quality of results is demonstrated by Nash–Sutcliffe and correlation coefficients that reach 0.82 and 0.92, respectively, in discharge comparison. With water levels, the obtained coefficients are lower but the quality of results still remains high; Nash–Sutcliffe and correlation coefficients reach 0.77 and 0.89, respectively, in the upstream part of the catchment. This analysis demonstrates the performance of the deterministic distributed modeling approach in simulating hydrological processes one more time; it also confirms the usefulness of this model with ungauged catchments or large catchments. Additionally, this analysis proves the role of multi-calibration in increasing the accuracy of hydrological models for large catchments.

scholarly journals Real-Time Analysis and Forecasting of Multisite River Flow Using a Distributed Hydrological Model

2014 ◽  
Vol 2014 ◽  
pp. 1-9
Author(s):  
Mingdong Sun ◽  
Gwangseob Kim
Keyword(s):  
Real Time ◽  
River Flow ◽  
Hydrological Model ◽  
Dynamic State ◽  
Distributed Hydrological Model ◽  
Time Analysis ◽  
Radar Rainfall ◽  
Real Time Analysis ◽  
Nakdong River Basin ◽  
Time Calibration

A spatial distributed hydrological forecasting system was developed to promote the analysis of river flow dynamic state in a large basin. The research presented the real-time analysis and forecasting of multisite river flow in the Nakdong River Basin using a distributed hydrological model with radar rainfall forecast data. A real-time calibration algorithm of hydrological distributed model was proposed to investigate the particular relationship between the water storage and basin discharge. Demonstrate the approach of simulating multisite river flow using a distributed hydrological model couple with real-time calibration and forecasting of multisite river flow with radar rainfall forecasts data. The hydrographs and results exhibit that calibrated flow simulations are very approximate to the flow observation at all sites and the accuracy of forecasting flow is gradually decreased with lead times extending from 1 hr to 3 hrs. The flow forecasts are lower than the flow observation which is likely caused by the low estimation of radar rainfall forecasts. The research has well demonstrated that the distributed hydrological model is readily applicable for multisite real-time river flow analysis and forecasting in a large basin.

Sign in / Sign up

Export Citation Format

Share Document