Weather Radar and Hydrology Edited by

Key words flood forecast; raingauge; radar

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 478-483

Comparison of raingauge and NEXRAD radar rainfall data for streamflow simulation for a southern Ontario catchment
Rohit Sharma¹, Ramesh Rudra², Syed Ahmed² &
Bahram Gharabaghi²

1 Water Resources Analyst, Calder Engineering Ltd, Bolton, Ontario L7E 3B2, Canada

2 School of Engineering, University of Guelph, Guelph, Ontario N1G 2W1, Canada

[e-mail]
Abstract The aim of this paper is to compare simulated flows using radar rainfall inputs with those obtained using raingauge rainfall. Both versions of simulated flows are also compared with the observed streamflow. The differences in rainfall volume and spatial variability of these datasets were evaluated for 10 storm events at hourly and daily time-scales using the Hydrologic Engineering Center’s Hydrologic Modelling System (HEC-HMS). The model was run in event-mode using the SCS Curve Number method and the Green and Ampt Infiltration method for a catchment in southern Ontario, Canada. For most of the events, the runoff hydrographs obtained using raingauge rainfall had better correlation with observed flows than those obtained using radar rainfall, or merged rainfall inputs. However, the merged rainfall gave better runoff simulations than those obtained using only-radar rainfall data. The use of “only” radar rainfall for hydrological modelling resulted in erroneous outputs. Therefore, adjustment of radar rainfall is important prior to its use for runoff simulations. The estimation of antecedent catchment conditions played a dominant role in the event simulations; therefore, the initial parameters should be carefully selected and calibrated. The SCS Curve Number option gave relatively better results in terms of runoff amount, peak flow-rate, and time to peak, than those obtained using the Green-Ampt Infiltration option.

Key words radar hydrology; hydrological modelling; HEC-HMS; heavy rainfall events

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 484-489.

Potential of radar data for flood forecasting and warning in lowland catchments in Ireland
M. B. DESTA, F. O’LOUGHLIN & M. BRUEN

School of Architecture, Landscape and Civil Engineering, University College Dublin, and
Centre for Water Resources Research, Belfield, Newstead Building, Dublin 4, Ireland

[e-mail]
Abstract This paper describes the development of a radar rainfall forecasting method and its use for flood forecasting using a data stream from Met Éireann’s radar at Dublin Airport. It is applied to four relatively flat catchments of different sizes on the eastern side of Ireland. The first objective was to determine the value of the radar precipitation information for hydrological applications in general, and the second was to assess if there is added value in applying Quantitative Precipitation Forecasting (QPF). A TREC-type procedure was used to generate QPF. The precipitation estimates are compared to contemporaneous raingauge measurements and the discharge estimates are compared to measured river flows. Preliminary results suggest that, with a 15-min radar cycle, this extends the acceptable performance by only an additional 1 h of lead time. While this is significant for the smaller catchments, it is less so for catchments with longer lag times.

Key words radar; rainfall; flood forecasting; rainfall forecasting; QPF

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012),490-495.

Operational use of nowcasting methods for hydrological forecasting by the Czech Hydrometeorological Institute
Lucie BŘezková¹, petr novák², milan Šálek¹, hana kyznarová², martin jonov³, petr frolík² & zbyněk sokol⁴

1 Czech Hydrometeorological Institute, Kroftova 43, 616 67 Brno, Czech Republic

[e-mail]

2 Czech Hydrometeorological Institute, Na Šabatce 17, 143 06 Prague, Czech Republic

3 Czech Hydrometeorological Institute, K myslivně 3/2182, 708 Ostrava, Czech Republic

4 Institute of Atmospheric Physics ASCR, Bocni II, 1401, 141 31 Prague, Czech Republic
Abstract The Czech Hydrometeorological Institute is the primary agency responsible for monitoring and forecasting of river stages at national level. In recent years, precipitation estimation and nowcasting tools derived from radar data were established. Together with hydrological models, these tools were tested for use in flash flood forecasting. The high uncertainty of predicting such a type of phenomena leads to using various nowcasting methods for estimation of predicted rainfall totals. This “variant-approach” was tested on case studies and is going to be set up for operational testing for pilot catchments. Detailed case studies of two extreme flash floods, which occurred on 24 June 2006, are presented and serve to demonstrate the possibilities and limitations of this method.

Key words flash flood; rainfall–runoff model; hydrological forecast; weather radar; heavy precipitation

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 496-501.

Flood nowcasting in the southern Swiss Alps using radar ensemble
Katharina Liechti¹, Felix Fundel¹, urs Germann² & Massimiliano Zappa¹

1 Swiss Federal Research Institute WSL, Zürcherstrasse 111, CH-8903 Birmensdorf, Switzerland

[e-mail]

2 Federal Office for Meteorology and Climatology, MeteoSwiss, Via Monti 146, CH-6605 Locarno, Switzerland
Abstract Since April 2007 the MeteoSwiss radar ensemble product REAL has been in operation and used for operational flash flood nowcasting by the WSL. REAL is computed for an area in the southern Swiss Alps where orographic and convective precipitation is frequent. These ensemble QPEs are processed by the semi-distributed hydrological model PREVAH. This provides operational ensemble nowcasts for several basins with areas from 44 to 1500 km² prone to flash floods and floods, respectively. Performances of discharge nowcasts driven by REAL are compared to performances of nowcasts forced by deterministic radar QPE and by interpolated raingauge data. We show that REAL outperforms deterministic radar QPE over the whole range of discharges, while the intercomparison with interpolated raingauge data is threshold dependent. Further we show that even though REAL nowcasts are underdispersive they have skill and can be a valuable means to produce hydrological nowcasts especially in ungauged catchments.

Key words radar ensemble; nowcasting; flash flood; flood; probabilistic; verification

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 502-507

Assessment of typhoon flood forecasting accuracy for various quantitative precipitation estimation methods
Tsung-Yi Pan¹, Yong-Jun Lin¹, Tsang-Jung Chang^1,2,3,
Jihn-Sung Lai^1,2,4 & Yih-Chi Tan^1,2

1 Center for Weather Climate and Disaster Research, 2 Department of Bioenvironmental System Engineering, 3 Ecological Engineering Research Center, 4 Hydrotech Research Institute, National Taiwan University, Taipei, Taiwan, China

[e-mail]
Abstract The main objective of the study is to obtain reliable rainfall estimates using raingauge and radar data. Different quantitative precipitation estimation (QPE) methods are tested and discussed, including (1) using kriging interpolation employing all raingauge data; (2) using radar products based on radar-reflectivity vs rain-rate (Z-R) formula; (3) using radar products adjusted by all raingauge data; and (4) using radar products adjusted by data from essential raingauges through network optimization with kriging. The estimated rainfalls are employed as the inputs for rainfall–runoff modelling. It is found that the QPE using radar products adjusted by all raingauge data provides superior performance. In addition, we found the estimation method using radar products adjusted by data from essential raingauges through network optimization with kriging can not only provide satisfactory results with efficiency for the spatial heterogeneity of rainfall distributions, but also simplify the raingauge network, reducing maintenance costs.

Key words quantitative precipitation; kriging interpolation; radar; raingauge; flood forecasting accuracy

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 508-513.

Ensemble nowcasting of river discharge by using radar data: operational issues on small- and medium-size basins
F. Silvestro & N. Rebora

CIMA research foundation, Savona, Italy

[e-mail]
Abstract Many efforts have been made in order to improve the reliability of quantitative precipitation estimation and to use radar data to forecast future rainfall evolution through nowcasting systems. From this perspective the use of stochastic nowcasting algorithms plays a key role both for taking into account the uncertainty associated with the prediction of rainfall and for generation of possible short-term evolution of the precipitation field. Propagation of the uncertainty to ground effects by using a rainfall–runoff model is a further step to completely exploit the weather radar systems when forecasting the consequences of severe events. We created a nowcasting chain for generating discharge scenarios based on the following procedures: (1) an algorithm for observed rainfall estimation; (2) an algorithm for probabilistic nowcasting (PhaSt); and (3) a distributed hydrological model (DRiFt). Some examples of application in an operational context on small-and medium-sized basins are presented

Key words discharge; flood; nowcasting; ensemble; probabilistic

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 514-519.

Influence of rainfall spatial variability on hydrological modelling: study by simulations
I. EMMANUEL¹, H. ANDRIEU², E. LEBLOIS³ & N. JANEY⁴

1 PRES L’UNAM, Ifsttar, Département GER, CS4, 44341 Bouguenais, France

[e-mail]

2 PRES L’UNAM, Ifsttar, Département GER and IRSTV FR CNRS 2488 Bouguenais, France

3 CEMAGREF, 3 B Quai Chauveau, 69009 Lyon, France

4 LIFC, UFR Sciences et Techniques, 16 route de Gray, 25030 Besançon, France
Abstract This work presents a simulation chain which enables studying the significance of rainfall spatial variability on flood runoff. A turning-band-method rainfall generator is used to simulate rainfall fields of different space–time variability. Catchments are extracted from Diffusion-Limited Aggregation structures. Three different rainfall–runoff models are implemented and the Hayami function is used to propagate runoff. Two spatial rainfall resolutions are taken into account: 250  250 m² and the rainfall average over the catchment. In this context hydrological studies are carried out. The influence of size of the catchment, of its production function and its response time are analysed. Hydrographs are compared in order to determine the interest for hydrology of detailed knowledge of rainfall. This work is currently ongoing.

Key words rainfall generator; catchment simulator; spatial variability; hydrological modelling

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012)., 520-525

Quantifying catchment-scale storm motion and its effects on flood response
DAVIDE ZOCCATELLI¹ , MARCO BORGA¹,
Efthymios I. nIkolopoulos¹ & Emmanouil N. Anagnostou²

1 Department of Land and Agroforest Environments, University of Padova, Legnaro (PD), Italy

[e-mail]

2 Department of Civil and Environmental Engineering, University of Connecticut, Storrs, USA
Abstract We introduce the concept of catchment-scale storm velocity and illustrate its evaluation for a flash flood case study. The computation of the catchment-scale storm velocity takes into account the overall dynamics of the storm motion over the catchment, reflecting the filtering effect of the catchment morphology with respect to the storm kinematics. Catchment-scale storm velocity is quantified for the
29 August 2003 extreme storm that occurred on the 600 km² Fella basin in the eastern Italian Alps. A spatially distributed rainfall–runoff model is used to evaluate the effects of the storm velocity on flood modelling for four sub-basins. It is shown that storm velocity exhibits rather moderate values, in spite of the strong kinematic characteristics of individual storm elements. Consistent with this observation, hydrologic simulations show that storm motion has an almost negligible effect on the flood response modelling.

Key words weather radar; flash flood; space-time rainfall variability; storm motion

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 526-531

Improvement of rainfall–runoff modelling with distributed radar rainfall data: a case study in the Lez, French Mediterranean, catchment
M. Coustau, V. Borrell-Estupina & C. Bouvier

Hydrosciences Montpellier (UMR 5569 CNRS-IRD-UM), 300 avenue du Pr. Emile Jeanbrau, 34 000 Montpellier, France

[e-mail]
Abstract The Mediterranean catchments in the south of France are prone to intense rainfall leading to destructive flash floods. These rainfalls mainly occur in autumn and show a high spatial variability. This study aims to assess the quality and impact in hydrological modelling of the radar rainfall data, in the Lez catchment (114 km²) near Montpellier, France. Comparison of both the raingauges and radar data proved to be satisfactory for events at the beginning of autumn. In contrast, important differences appeared for events occurring at the end of autumn. This can be explained by the weak vertical extension of the clouds and the low altitude of the 0°C isotherm in this period, which could affect the accuracy of radar measurements due to the distance between the basin and the radar (~60 km). To take advantage of the spatial variability of the radar rainfall data, the flood simulations were performed through a distributed event-based rainfall–runoff model. The model was calibrated using a sample of 21 floods observed from 1994 to 2008 where both recording raingauge and radar rainfall data were available. When the radar rainfalls were reliable, they led to: (i) an improvement of the optimal flood simulation at the outlet, and (ii) an improvement of the relationship between the calibrated initial condition of the model and external predictors such as piezometric level, baseflow and Hu2 index from the Meteo-France SIM model. Installation of an X-band radar near the study area could improve rainfall estimation at the end of the autumn for the Lez catchment and the Montpellier agglomeration.

Key words flash flood; distributed rainfall–runoff model; event-based model; radar rainfall

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 532-537

Representing the spatial variability of rainfall for input to the G2G distributed flood forecasting model: operational experience from the Flood Forecasting Centre
david price¹, charlie pilling¹, GaviN Robbins¹, andy lane¹, graeme boyce¹, keith fenwick¹, RobERT J. moore²,
joanne COLES³, tim harrison³ & marc van Dijk⁴

1 Flood Forecasting Centre, Met Office, FitzRoy Road, Exeter, Devon EX1 3PB, UK

[e-mail]

2 Centre for Ecology & Hydrology, Wallingford OX10 8BB, UK

3 Environment Agency, Horizon House, Deanery Road, Bristol BS1 5AH, UK

4 Deltares, Inland Water Systems, Deltares, Delft, The Netherlands
Abstract Over the year 2010 the Flood Forecasting Centre (FFC) calibrated and implemented a distributed flood forecasting model to support the FFC’s remit to provide flood risk forecasts across England and Wales, UK. The distributed nature of the model, designed to run at 15-min time-steps on a 1 km² grid, enables the spatial variability of rainfall measurements and forecasts, rather than lumped catchment averages, to be captured. Such a distributed model should therefore benefit greatly from the spatial and temporal resolution afforded by radar observations. Initial results have highlighted the importance of the quality of the gridded rainfall fields and in a number of cases erroneous radar rainfall data have been shown to contribute to poor model performance. It is suggested that gridded datasets of sufficient quality will be best provided by capturing the spatial variability inherent in radar data together with raingauge data in a merged product.

Key words flood forecasting; distributed flood forecasting model; Flood Forecasting Centre; radar; Grid-to-Grid model

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 538-543.

Countrywide flood forecasting in Scotland: challenges for hydrometeorological model uncertainty and prediction
MICHAEL CRANSTON¹, RICHARD MAXEY¹, AMY TAVENDALE¹,
PETER BUCHANAN², ALAN MOTION², STEVEN COLE³, ALICE ROBSON³, ROBERT J. MOORE³ & ALEX MINETT⁴

1 Scottish Environment Protection Agency, Flood Forecasting and Warning Section, 7 Whitefriars Crescent, Perth, UK

[e-mail]

2 Met Office, Operations Centre, Davidson House, Aberdeen Science and Technology Park, Aberdeen, UK

3 Centre for Ecology & Hydrology, Wallingford, UK

4 Deltares, Rotterdamsewag 185, Delft, The Netherlands
Abstract The Scottish Flood Forecasting Service, a new partnership between the Met Office and the Scottish Environment Protection Agency, aims to make best use of weather and river forecasting expertise in providing improved flood resilience and vigilance for emergency responders across Scotland. Flood guidance employs a blend of experience, professional assessment and input from meteorological and hydrological models. For countrywide forecasts, the CEH-developed Grid-to-Grid model is planned to be the key forecasting tool: it employs rainfall estimates from raingauges, radar and weather models to produce forecast river flows up to 5 days ahead on a 1-km grid across the Scottish mainland. Probabilistic flood forecasts, using ensemble rainfalls as input, are planned in a future phase. Use of rainfall as input to hydrological models is a challenge in Scotland, especially given the terrain and sparse radar and raingauge network coverage, and makes forecasting uncertain. However, the merged hydrological and meteorological capabilities of the new service bring tangible benefits for improved flood forecasting.

Key words flood; forecasting; hydrometeorology

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 544-549.

Distributed flood forecasting for the management of the road network in the Gard Region (France)
J.-P. NAULIN, E. Gaume & O. PAYRASTRE

French Institute of Science and Technology for Transport, Development and Networks, Centre de Nantes, France

[e-mail]
Abstract A prototype of a road submersion warning system, providing a rating of road submersion risks every 15 minutes during a storm event, at about 2000 points where roads and rivers intersect, has been developed for the Gard region (French Mediterranean area). The computed risks result from the confrontation between discharges produced by a distributed rainfall–runoff model and the estimated susceptibility to flooding of the intersection points. The warning system is validated against road inundations reported by the local road management service. The comparison of the performances of this framework fed with various quantitative precipitation estimates (1-km² grid interpolation of point rainfall measurements or radar products) is presented herein. This case study, based on highly distributed rainfall–runoff modelling and on a rich set of indirect observations of the flood magnitudes, provides a priori an ideal framework to evaluate the usefulness of weather radar products for hydrological applications.

Key words flash flood; radar; road network; rainfall–runoff; ungauged

Weather Radar and Hydrology

(Proceedings of a symposium held in Exeter, UK, April 2011) (IAHS Publ. 351, 2012), 550-555

The AIGA method: an operational method using radar rainfall for flood warning in the south of France
PIERRE Javelle¹, JEAN Pansu², PATRICK Arnaud¹, YVES Bidet² & BRUNO Janet³

1 Cemagref, Centre Régional d'Aix-en-Provence, CS 40061, 13182 AIX EN PROVENCE Cedex 5, France

[e-mail]

2 Météo-France, Direction Interrégionale Sud-Est, 2, Bd Chateau-Double, 13098 Aix-en-Provence cedex 02, France

3 Ministère de l’Ecologie, du développement durable des transports et du logement (MEDDTL), Direction gérérale de la prévention des risques (DGPR) Service des risques naturels et hydrauliques (SRNH), Service Central d'Hydrométéorologie et d'Appui à  la Prévision des Inondations (SCHAPI), 42, avenue Gaspard Coriolis 31057, Toulouse Cedex 01, France