A paleo‐hydro‐geomorphological perspective on urban flood risk assessment

Comprehensive flood risk assessment requires enhanced understanding of the coevolution of the river and its floodplain occupation. Paleoflood analysis to determine flood prone areas in combination with numerical simulations to estimate flood hazard and a historical analysis of urban development to consider the evolution of exposure to floods is a possible way forward. The well‐documented 2006 extreme flood in the Biobío River system and the impacted metropolitan area of Concepción, Chile (~1 million inhabitants) was used as a complex scenario to test the reliability of the proposed method. Results showed that flood prone areas determined with hydro‐geomorphological methods are consistent with those computed with numerical models based on detailed digital elevation models. The flood generation via superficial flow pathways resulting in inundated areas could explain that rivers tend to reactivate paleochannels in extreme conditions. Urban development progressively increased the city's exposure to floods from 0 ha in 1,751 to 1,363 ha in 2006 evidencing a lack of appropriate flood risk management. The 100‐year peak discharge resulted in a high flood risk for about 5% of the total urbanized area of Concepción, and higher discharges are likely to reactivate a paleochannel that crosses the current city centre. We conclude that the proposed paleo hydro‐geomorphology, hydraulic, and urban planning multimethod approach is a necessary tool to enhance understanding of flood risk in complex scenarios to improve flood risk management.     

A comparison of one‐ and two‐dimensional approaches to modelling flood inundation over complex upland floodplains

A much understudied aspect of flood inundation is examined, i.e. upland environments with topographically complex floodplains. Although the presence of high‐resolution topographic data (e.g. lidar) has improved the quality of river flood inundation predictions, the optimum dimensionality of hydraulic models for this purpose has yet to be fully evaluated for situations of both topographic and topological (i.e. the connectivity of floodplain features) complexity. In this paper, we present the comparison of three treatments of upland flood inundation using: (a) a one‐dimensional (1D) model (HEC‐RAS v. 3·1·2) with the domain defined as series of extended cross‐sections; (b) the same 1D model, but with the floodplain defined by a series of storage cells, hydraulically connected to the main river channel and other storage cells on the floodplain according to floodplain topological characteristics; (c) a two‐dimensional (2D) diffusion wave treatment, again with explicit representation of floodplain structural features. The necessary topographic and topological data were derived using lidar and Ordnance Survey Landline data. The three models were tested on a 6 km upland reach of the River Wharfe, UK. The models were assessed by comparison with measured inundation extent. The results showed that both the extended cross‐section and the storage cell 1D modes were conceptually problematic. They also resulted in poorer model predictions, requiring incorrect parameterization of the main river to floodplain flux in order to approach anything like the level of agreement observed when the 2D diffusion wave treatment was assessed. We conclude that a coupled 1D–2D treatment is likely to provide the best modelling approach, with currently available technology, for complex floodplain configurations. Copyright © 2007 John Wiley & Sons, Ltd.     

Levee effects upon flood levels: an empirical assessment

This study used stream gauge records to assess the impact of levees on flood levels, providing an empirical test of theoretical and model predictions of the effects on local flood response. Focusing upon a study area in Illinois and Iowa for which levee records were available, we identified 203 gauges with ≥ 50 years hydrological record, including 15 gauges where a levee was constructed during the period of record. At these sites, step‐change analysis utilizing regression residuals tested levee‐related stage changes and levels of significance and quantified the magnitudes of stage changes. Despite large differences in stream sizes, levee alignments, and degree of floodplain constriction, the post‐levee rating‐curve adjustments showed consistent signatures. For all the study sites, stages for below bankfull (non‐flood) conditions were unaffected by levee construction. For above bankfull (flood) conditions, stages at sites downstream of their associated levees also were statistically indistinguishable before versus after levee construction. However, at all sites upstream of levees or within leveed reaches, stages increased for above bankfull conditions. These increases were abrupt, statistically significant, and generally large in magnitude – ranging up to 2.3 m (Wabash River at Mt. Carmel, IL). Stage increases began when discharge increased above bankfull flow and generally increased in magnitude with discharge until the associated levee(s) were overtopped. Detailed site assessments and supplementary data available from some sites helped document the dominant mechanisms by which levees can increase flood levels. Levee construction reduces the area of the floodplain open to storage of flood waters and reduces the width of the floodplain open to conveyance of flood flow. Floodplain conveyance often is underestimated or ignored, but Acoustic Doppler Current Profiler (ADCP) measurements analysed here confirm previous studies that up to 70% or more of the total discharge during large floods (~3% chance flood) can move over the floodplain. Upstream of levees and levee‐related floodplain constriction, backwater effects reduce flow velocities relative to pre‐levee conditions and, thus, increase stages for a given discharge. The empirical results here confirm a variety of theoretical predictions of levee effects but suggest that many one‐dimensional model‐based predictions of levee‐related stage changes may underestimate actual levee impacts. Copyright © 2011 John Wiley & Sons, Ltd.     

Interactions between subgrid‐scale resolution,feature representation and grid‐scale resolution in flood inundation modelling

Numerical modelling of flood inundation over large and complex floodplains often requires mesh resolutions coarser than the structural features (e.g. buildings) that are known to influence the inundation process. Recent research has shown that this mismatch is not well represented by conventional roughness treatments, but that finer‐scale features can be represented through porosity‐based subgrid‐scale treatments. This paper develops this work by testing the interactions between feature representation, subgrid‐scale resolution and mesh resolution. It uses as the basis for this testing a 2D diffusion‐based flood inundation model which is applied to a 2004 flood event in a topologically complex upland floodplain in northern England. This study formulated simulations with different grid mesh resolution and subgrid mesh ratio. The sensitivity of the model to mesh resolution and roughness specification was investigated. Model validation and verification suggest that the subgrid treatment with higher subgrid mesh ratio can give much improved predictions of flood propagation, in particular, in terms of the predicted water depth. This study also highlighted the limitation of using at‐a‐point in time inundation extent for validation of flood models of this type. Copyright © 2010 John Wiley & Sons, Ltd.     

Morphodynamics of a 1000‐year flood in the Kamp River,Austria, and impacts on floodplain morphology

This paper investigates the impact of a 1000‐year flood in August 2002 on floodplains and valley morphology of an Austrian mixed alluvial bed rock river. Discharges with a recurrence interval between 500 and 2000 years caused distinctive overbank scouring and material deposition in the floodplains. After the 1000‐year flood, those morphologically affected areas were at random intervals documented over the whole longitudinal profile. In addition to overbank erosion in curved sections (cut‐offs), the river bed locally widened, floodplain stripping occurred and local overbank scours were documented along straight parts of the river. A hydrodynamic‐numerical model, combined with field measurements, was used to analyse the cause of these erosional landforms. Based on the modelled hydraulic conditions for a one‐year flood (30–78 m³ s^–1) and the catastrophic 2002 event (700–800 m³ s^–1), the numerical results allowed a cause‐effect study with 19 parameters. Deterministic and statistical analysis (ANOVA, discriminant analysis) showed that the morphodynamic effects of the 2002 flood were influenced by the variability of valley morphology of the Kamp River, which led partially to supercritical flow during flood constriction. These processes were in some cases also anthropogenically influenced. Lateral constriction and expansion of the valley geometry over short distances led to scouring and aggradation within the inundated areas during the event. These morphological features were therefore responsible for the elongated scour holes in the floodplains. Copyright © 2009 John Wiley & Sons, Ltd.     

A fast simplified model for predicting river flood inundation probabilities in poorly gauged areas

The capability of a simple kinematic‐storage model (KSM) is analysed to be used as a tool for a Decision Support System for the evaluation of probability inundation maps in near real time in poorly gauged areas. KSM simulates the floodplain as a storage and assumes no exchange of momentum with the channel. For the in‐bank flow, the storage is modified through a coefficient for taking the variations of channel cross sections into account. The generalized likelihood uncertainty estimation approach is used for addressing the probability flood maps along with their associated uncertainties. The model is tested on two river reaches along the Tiber River in Central Italy where observed inundation maps are available for two recent flood events. Despite the inherent uncertainties present in the input data and in the model structure, the results show that the model reproduces reasonably well, in terms of both precision and accuracy, the observed inundated areas. Tests were performed at different digital elevation model resolutions, showing a small effect of the geometry on the maximum performance obtained. The very low computational times, the parsimony of the model and its low sensitivity to the quality of the geometry representation of the channel and the floodplain makes KSM very appealing for flood forecasting and early warning system applications in poorly gauged and inaccessible areas. Copyright © 2014 John Wiley & Sons, Ltd.     

Developing a hybrid multi‐model for peak flood forecasting

The present study aims to develop a hybrid multi‐model using the soft computing approach. The model is a combination of a fuzzy logic, artificial neural network (ANN) and genetic algorithm (GA). While neural networks are low‐level computational structures that perform well dealing with raw data, fuzzy logic deal with reasoning on a higher level by using linguistic information acquired from domain experts. However, fuzzy systems lack the ability to learn and cannot adjust themselves to a new environment. Moreover, experts occasionally make mistakes and thus some rules used in a system may be false. A network type structure of the present hybrid model is a multi‐layer feed‐forward network, the main part is a fuzzy system based on the first‐order Sugeno fuzzy model with a fuzzification and a defuzzification processes. The consequent parameters are determined by least square method. The back‐propagation is applied to adjust weights of network. Then, the antecedent parameters of the membership function are updated accordingly by the gradient descent method. The GA was applied to select the fuzzy rule. The hybrid multi‐model was used to forecast the flood level at Chiang Mai (under the big flood 2005) and the Koriyama flood (2003) in Japan. The forecasting results are evaluated using standard global goodness of fit statistic, efficient index (EI), the root mean square error (RMSE) and the peak flood error. Moreover, the results are compared to the results of a neuro‐genetic model (NGO) and ANFIS model using the same input and output variables. It was found that the hybrid multi‐model can be used successfully with an efficiency index (EI) more than 0·95 (for Chiang Mai flood up to 12 h ahead forecasting) and more than 0·90 (for Koriyama flood up to 8 h ahead forecasting). In general, all of three models can predict the water level with satisfactory results. However, the hybrid model gave the best flood peak estimation among the three models. Therefore, the use of fuzzy rule base, which is selected by GA in the hybrid multi‐model helps to improve the accuracy of flood peak. Copyright © 2009 John Wiley & Sons, Ltd.     

Application of tank,NAM, ARMA and neural network models to flood forecasting

Two lumped conceptual hydrological models, namely tank and NAM and a neural network model are applied to flood forecasting in two river basins in Thailand, the Wichianburi on the Pasak River and the Tha Wang Pha on the Nan River using the flood forecasting procedure developed in this study. The tank and NAM models were calibrated and verified and found to give similar results. The results were found to improve significantly by coupling stochastic and deterministic models (tank and NAM) for updating forecast output. The neural network (NN) model was compared with the tank and NAM models. The NN model does not require knowledge of catchment characteristics and internal hydrological processes. The training process or calibration is relatively simple and less time consuming compared with the extensive calibration effort required by the tank and NAM models. The NN model gives good forecasts based on available rainfall, evaporation and runoff data. The black‐box nature of the NN model and the need for selecting parameters based on trial and error or rule‐of‐thumb, however, characterizes its inherent weakness. The performance of the three models was evaluated statistically. Copyright © 2000 John Wiley & Sons, Ltd.     

Enforced self‐organizing map neural networks for river flood forecasting

Self‐organizing maps (SOMs) have been successfully accepted widely in science and engineering problems; not only are their results unbiased, but they can also be visualized. In this study, we propose an enforced SOM (ESOM) coupled with a linear regression output layer for flood forecasting. The ESOM re‐executes a few extra training patterns, e.g. the peak flow, as recycling input data increases the mapping space of peak flow in the topological structure of SOM, and the weighted sum of the extended output layer of the network improves the accuracy of forecasting peak flow. We have investigated an ESOM neural network by using the flood data of the Da‐Chia River, Taiwan, and evaluated its performance based on the results obtained from a commonly used back‐propagation neural network. The results demonstrate that the ESOM neural network has great efficiency for clustering, especially for the peak flow, and super capability of modelling the flood forecast. The topology maps created from the ESOM are interesting and informative. Copyright © 2007 John Wiley & Sons, Ltd.     

Urban fluvial flood modelling using a two‐dimensional diffusion‐wave treatment,part 1: mesh resolution effects

High‐resolution data obtained from airborne remote sensing is increasing opportunities for representation of small‐scale structural elements (e.g. walls, buildings) in complex floodplain systems using two‐dimensional (2D) models of flood inundation. At the same time, 2D inundation models have been developed and shown to provide good predictions of flood inundation extent, with respect to both full solution of the depth‐averaged Navier–Stokes equations and simplified diffusion‐wave models. However, these models have yet to be applied extensively to urban areas. This paper applies a 2D raster‐based diffusion‐wave model to determine patterns of fluvial flood inundation in urban areas using high‐resolution topographic data and explores the effects of spatial resolution upon estimated inundation extent and flow routing process. Model response shows that even relatively small changes in model resolution have considerable effects on the predicted inundation extent and the timing of flood inundation. Timing sensitivity would be expected, given the relatively poor representation of inertial processes in a diffusion‐wave model. Sensitivity to inundation extent is more surprising, but is associated with: (1) the smoothing effect of mesh coarsening upon input topographical data; (2) poorer representation of both cell blockage and surface routing processes as the mesh is coarsened, where the flow routing is especially complex; and (3) the effects of (1) and (2) upon water levels and velocities, which in turn determine which parts of the floodplain the flow can actually travel to. It is shown that the combined effects of wetting and roughness parameters can compensate in part for a coarser mesh resolution. However, the coarser the resolution, the poorer the ability to control the inundation process, as these parameters not only affect the speed, but also the direction of wetting. Thus, high‐resolution data will need to be coupled to a more sophisticated representation of the inundation process in order to obtain effective predictions of flood inundation extent. This is explored in a companion paper. Copyright © 2005 John Wiley & Sons, Ltd.     

Interactions between sediment delivery,channel change,climate change and flood risk in a temperate upland environment

This paper uses numerical simulation of flood inundation based on a coupled one‐dimensional–two‐dimensional treatment to explore the impacts upon flood extent of both long‐term climate changes, predicted to the 2050s and 2080s, and short‐term river channel changes in response to sediment delivery, for a temperate upland gravel‐bed river. Results show that 16 months of measured in‐channel sedimentation in an upland gravel‐bed river cause about half of the increase in inundation extent that was simulated to arise from climate change. Consideration of the joint impacts of climate change and sedimentation emphasized the non‐linear nature of system response, and the possibly severe and synergistic effects that come from combined direct effects of climate change and sediment delivery. Such effects are likely to be exacerbated further as a result of the impacts of climate change upon coarse sediment delivery. In generic terms, these processes are commonly overlooked in flood risk mapping exercises and are likely to be important in any river system where there are high rates of sediment delivery and long‐term transfer of sediment to floodplain storage (i.e. alluviation involving active channel aggradation and migration). Similarly, attempts to reduce channel migration through river bank stabilization are likely to exacerbate this process as without bank erosion, channel capacity cannot be maintained. Finally, many flood risk mapping studies rely upon calibration based upon combining contemporary bed surveys with historical flood outlines, and this will lead to underestimation of the magnitude and frequency of floodplain inundation in an aggrading system for a flood of a given magnitude. Copyright © 2006 John Wiley & Sons, Ltd.     

The effects of river restoration on catchment scale flood risk and flood hydrology

A rising exposure to flood risk is a predicted consequence of increased development in vulnerable areas and an increase in the frequency of extreme weather events due to climate change. In the face of this challenge, a continued reliance on engineered at‐a‐point flood defences is seen as both unrealistic and undesirable. The contribution of ‘soft engineering’ solutions (e.g. riparian forests, wood in rivers) to integrated, catchment scale flood risk management has been demonstrated at small scales but not larger ones. In this study we use reduced complexity hydrological modelling to analyse the effects of land use and channel changes resulting from river restoration upon flood flows at the catchment scale. Results show short sections of river‐floodplain restoration using engineered logjams, typical of many current restoration schemes, have highly variable impacts on catchment‐scale flood peak magnitude and so need to be used with caution as a flood management solution. Forested floodplains have a more general impact upon flood hydrology, with areas in the middle and upper catchment tending to show reductions in peak magnitude at the catchment outflow. The most promising restoration scenarios for flood risk management are for riparian forest restoration at the sub‐catchment scale, representing 20–40% of the total catchment area, where reductions in peak magnitude of up to 19% are observed through de‐synchronization of the timings of sub‐catchment flood waves. Sub‐catchment floodplain forest restoration over 10–15% of total catchment area can lead to reductions in peak magnitude of 6% at 25 years post‐restoration. Copyright © 2016 John Wiley & Sons, Ltd.     

Application of integrated back‐propagation network and self organizing map for flood forecasting

Much of the nonlinearity and uncertainty regarding the flood process is because hydrologic data required for estimation are often tremendously difficult to obtain. This study employed a back‐propagation network (BPN) as the main structure in flood forecasting to learn and to demonstrate the sophisticated nonlinear mapping relationship. However, a deterministic BPN model implies high uncertainty and poor consistency for verification work even when the learning performance is satisfactory for flood forecasting. Therefore, a novel procedure was proposed in this investigation which integrates linear transfer function (LTF) and self‐organizing map (SOM) to efficiently determine the intervals of weights and biases of a flood forecasting neural network to avoid the above problems. A SOM network with classification ability was applied to the solutions and parameters of the BPN model in the learning stage, to classify the network parameter rules and to obtain the winning parameters. The outcomes from the previous stage were then used as the ranges of the parameters in the recall stage. Finally, a case study was carried out in Wu‐Shi basin to demonstrate the effectiveness of the proposal. Copyright © 2009 John Wiley & Sons, Ltd.     

Floodplain stratigraphy of the ice jam dominated middle Yukon River: a new approach to long‐term flood frequency

Floodplain stratigraphy is used as a new method for reconstructing ice jam flood histories of northern rivers. The method, based on reconstruction of the sedimentary record of vertically‐accreting floodplains, relies on stratigraphic logging and interpretation of floodplain sediments, which result from successive ice jam floods, and radiocarbon dating of inter‐flood organic material for chronology. In a case study along a reach of the Yukon River that straddles the Yukon–Alaska border, the method is used to develop a record of ice jam flooding for the last 2000 years. Detailed chronostratigraphic logs from three sites along the Yukon River indicates that the long‐term recurrence interval varies depending on location, but ranges from approximately once in 25 years to once in 38 years (or a probability of ca 3–4% in any given year). This is broadly similar to the 4·5% probability of recurrence calculated from archival and gauged data at Dawson City, Yukon Territory, for the period 1898–2006. Two of the three study locations, with sufficient chronology, suggest a decrease in flood frequency in the last several hundred years relative to the preceding period at each site, broadly corresponding to the Little Ice Age, suggesting climate exerts some control over long‐term ice jam flood frequency. This study demonstrates that the floodplain sedimentary record offers the potential to extend records of ice jam flooding in remote, ungauged northern rivers and provides a broader temporal context for assessing the frequency and variability of ice jam flooding. Copyright © 2008 John Wiley & Sons, Ltd.     

Uncertainty in floodplain delineation: expression of flood hazard and risk in a Gulf Coast watershed

This paper investigates the development of flood hazard and flood risk delineations that account for uncertainty as improvements to standard floodplain maps for coastal watersheds. Current regulatory floodplain maps for the Gulf Coastal United States present 1% flood hazards as polygon features developed using deterministic, steady‐state models that do not consider data uncertainty or natural variability of input parameters. Using the techniques presented here, a standard binary deterministic floodplain delineation is replaced with a flood inundation map showing the underlying flood hazard structure. Additionally, the hazard uncertainty is further transformed to show flood risk as a spatially distributed probable flood depth using concepts familiar to practicing engineers and software tools accepted and understood by regulators. A case study of the proposed hazard and risk assessment methodology is presented for a Gulf Coast watershed, which suggests that storm duration and stage boundary conditions are important variable parameters, whereas rainfall distribution, storm movement, and roughness coefficients contribute less variability. The floodplain with uncertainty for this coastal watershed showed the highest variability in the tidally influenced reaches and showed little variability in the inland riverine reaches. Additionally, comparison of flood hazard maps to flood risk maps shows that they are not directly correlated, as areas of high hazard do not always represent high risk. Copyright © 2012 John Wiley & Sons, Ltd.     

Fuzzy computing based rainfall–runoff model for real time flood forecasting

This paper analyses the skills of fuzzy computing based rainfall–runoff model in real time flood forecasting. The potential of fuzzy computing has been demonstrated by developing a model for forecasting the river flow of Narmada basin in India. This work has demonstrated that fuzzy models can take advantage of their capability to simulate the unknown relationships between a set of relevant hydrological data such as rainfall and river flow. Many combinations of input variables were presented to the model with varying structures as a sensitivity study to verify the conclusions about the coherence between precipitation, upstream runoff and total watershed runoff. The most appropriate set of input variables was determined, and the study suggests that the river flow of Narmada behaves more like an autoregressive process. As the precipitation is weighted only a little by the model, the last time‐steps of measured runoff are dominating the forecast. Thus a forecast based on expected rainfall becomes very inaccurate. Although good results for one‐step‐ahead forecasts are received, the accuracy deteriorates as the lead time increases. Using the one‐step‐ahead forecast model recursively to predict flows at higher lead time, however, produces better results as opposed to different independent fuzzy models to forecast flows at various lead times. Copyright © 2004 John Wiley & Sons, Ltd.     

Urban fluvial flood modelling using a two‐dimensional diffusion‐wave treatment,part 2: development of a sub‐grid‐scale treatment

This paper develops and tests a sub‐grid‐scale wetting and drying correction for use with two‐dimensional diffusion‐wave models of urban flood inundation. The method recognizes explicitly that representations of sub‐grid‐scale topography using roughness parameters will provide an inadequate representation of the effects of structural elements on the floodplain (e.g. buildings, walls), as such elements not only act as momentum sinks, but also have mass blockage effects. The latter may dominate, especially in structurally complex urban areas. The approach developed uses high‐resolution topographic data to develop explicit parameterization of sub‐grid‐scale topographic variability to represent both the volume of a grid cell that can be occupied by the flow and the effect of that variability upon the timing and direction of the lateral fluxes. This approach is found to give significantly better prediction of fluvial flood inundation in urban areas than traditional calibration of sub‐grid‐scale effects using Manning's n. In particular, it simultaneously reduces the need to use exceptionally high values of n to represent the effects of using a coarser mesh process representation and increases the sensitivity of model predictions to variation in n. Copyright © 2005 John Wiley & Sons, Ltd.     

Defining the floodplain in hydrologically‐variable settings: implications for flood risk management

Flood risk management is an essential responsibility of state governments and local councils to ensure the protection of people residing on floodplains. Globally, floodplains are under increasing pressure from growing populations. Typically, the engineering‐type solutions that are used to predict local flood magnitude and frequency based on limited gauging data are inadequate, especially in settings which experience high hydrological variability. This study highlights the importance of incorporating geomorphological understanding into flood risk management in southeast Queensland (SEQ), an area badly affected by extreme flood events in 2011 and 2013. The major aim of this study is to outline the hydrological and sedimentological characteristics of various ‘inundation surfaces’ that are typical of catchments in the sub‐tropics. It identifies four major inundation surfaces; within‐channel bench [Q ~ 2.33 yr average recurrence interval (ARI)]; genetic floodplain (Q = 20 yr ARI); hydraulic floodplain (20 yr < Q ≤ 200 yr ARI) and terrace (Q > 1000 yr ARI). These surfaces are considered typical of inundation areas within, and adjacent to, the large macrochannels common to this region and others of similar hydrological variability. An additional area within genetic floodplains was identified where flood surfaces coalesce and produce an abrupt reduction in channel capacity. This is referred to here as a Spill‐out Zone (SOZ). The associated vulnerability and risk of these surfaces is reviewed and recommendations made based on incorporating this geomorphological understanding into flood risk assessments. These recommendations recognize the importance to manage for risks associated with flow inundation and sediment erosion, delivery and deposition. The increasing availability of high resolution topographic data opens up the possibility of more rapid and spatially extensive assessments of key geomorphic processes which can readily be used to predict flood risk. Copyright © 2016 John Wiley & Sons, Ltd.     

Uncertainties in real‐time flood forecasting with neural networks

Although artificial neural networks (ANNs) have been applied in rainfall runoff modelling for many years, there are still many important issues unsolved that have prevented this powerful non‐linear tool from wide applications in operational flood forecasting activities. This paper describes three ANN configurations and it is found that a dedicated ANN for each lead‐time step has the best performance and a multiple output form has the worst result. The most popular form with multiple inputs and single output has the average performance. In comparison with a linear transfer function (TF) model, it is found that ANN models are uncompetitive against the TF model in short‐range predictions and should not be used in operational flood forecasting owing to their complicated calibration process. For longer range predictions, ANN models have an improved chance to perform better than the TF model; however, this is highly dependent on the training data arrangement and there are undesirable uncertainties involved, as demonstrated by bootstrap analysis in the study. To tackle the uncertainty issue, two novel approaches are proposed: distance analysis and response analysis. Instead of discarding the training data after the model's calibration, the data should be retained as an integral part of the model during its prediction stage and the uncertainty for each prediction could be judged in real time by measuring the distances against the training data. The response analysis is based on an extension of the traditional unit hydrograph concept and has a very useful potential to reveal the hydrological characteristics of ANN models, hence improving user confidence in using them in real time. Copyright © 2006 John Wiley & Sons, Ltd.     

Distributed rainfall‐runoff modelling for flood frequency estimation and flood forecasting

Nowadays, in the scientific literature many rainfall‐runoff (RR) models are available ranging from simpler ones, with a limited number of parameters, to highly complex ones, with many parameters. Therefore, the selection of the best structure and parameterisation for a model is not straightforward as it is dependent on a number of factors: climatic conditions, catchment characteristics, temporal and spatial resolution, model objectives, etc. In this study, the structure of a continuous semi‐distributed RR model, named MISDc (‘Modello Idrologico Semi‐Distribuito in continuo’) developed for flood simulation in the Upper Tiber River (central Italy) is presented. Most notably, the methodology employed to detect the more relevant processes involved in the modelling of high floods, and hence, to build the model structure and its parameters, is developed. For this purpose, an intense activity of monitoring soil moisture and runoff in experimental catchments was carried out allowing to derive a parsimonious and reliable continuous RR model operating at an hourly (or smaller) time scale. Specifically, in order to determine the catchment hydrological response, the important role of the antecedent wetness conditions is emphasized. The application of MISDc both for design flood estimation and for flood forecasting is reported here demonstrating its reliability and also its computational efficiency, another important factor in hydrological practice. As far as the flood forecasting applications are concerned, only the accuracy of the model in reproducing discharge hydrographs by assuming rainfall correctly known throughout the event is investigated indepth. In particular, the MISDc has been implemented in the framework of Civil Protection activities for the Upper Tiber River basin. Copyright © 2011 John Wiley & Sons, Ltd.