There were two principal aims to this research: (i) to gain a better understanding of the authoritarian response to the flood hazard; (ii) to produce a predictive model of the residential response to the problem. Following an initial review of flood plain management techniques, three scales of spatial analysis were identified. I. The National Level: A broad investigation was undertaken into flood plain management programmes in Britain. This survey illustrated the narrow authoritarian response in the past. and the recent move towards non-structural measures. particularly forecasting and warning schemes. and highlighted the need for a greater consideration of social factors in flood plain planning. 2. The Regional Level: Regional level studies examined various flood types the severity of the hazard. and the response to the flood problem by the responsible organisations in the county of Cumbria. 3. The Local Level: Detailed local level surveys were carried out at Carlisle and Appleby to assess the significance of residential and commercial behaviour in affecting the extent of flood losses. An extensive questionnaire survey of residents and business-men was undertaken in the two research centres, to examine the behavioural aspects of the flood plain population in terms of perception and awareness of the flood hazard, the degree of fear associated with flooding, the awareness of authoritarian alleviation measures, and the perceived effectiveness of individual adjustments to the problem. The evidence indicated that the perceived hazard is more important than the actual hazard in determining the individual response to the flood problem. The final research model suggested certain significant social characteristics, which could be used to predict flood plain behaviour and thus reduce potential flood losses. This is critical to flood loss reduction programmes, especially with the .trend towards non·structural alleviation schemes, since inefficient flood plain behaviour could significantly reduce the effectiveness of such programmes.

Knowledge of how local changes in runoff generation propagate through the stream network to create impacts downstream at the larger catchment scale is lacking. This issue needs to be urgently addressed, particularly as many organisations involved in flood risk management favour interventions that manage runoff at source. To enable research into propagation of catchment flood runoff and its impacts, a new modelling approach has been developed. The catchment is discretised into the component hillslopes/fields, which are connected to a dense dendritic stream network. Runoff generated from the landscape units is represented as packets of water, which are tracked through the stream network using a numerical solution of the non-inertial form of the Saint Venant equations. The packets carry information, including the volume, location and time of runoff generation. To recover the information, the packets travelling past particular downstream points in the network are interrogated. The recovered information can be used in many ways. For example, it is possible to create spatial maps, showing the source locations and intensities of runoff contributing to the catchment hydrograph peak, or to break down the hydrograph into contributing hydrographs, one per sub-catchment or per area undergoing change. The importance of this novel approach is that it explicitly links the local scale, at which landscape adaptation measures are implemented, and the catchment scale, where impact occurs. An application of this type is demonstrated for the Eden Catchment, UK, providing novel and informative visual representations of spatial runoff generation and propagation.

A deficiency in current understanding of flood response is the variation of flood generation at different spatial scales as a function of spatial and temporal variations in storm rainfall. There is a lack of studies showing the process of transformation from rainfall to stream flow at a range of consecutive scales. This process is complex and varies with catchment size. The aim of the project is therefore to investigate the relationship between rainfall spatial variability anc lood response through a multiscale nested experiment.

In the UK, perceived increases ih high impact flood events over the last decade and broad scale recognition of the enhanced flood risks associated with future climate change predictions, have reinforced the need for improved understanding and management of processes governing peak flow responses. This thesis investigates the effects of agricultural land uses on the hydrology of rural areas at a range of spatial and temporal scales. At the catchment scale, 48 catchments and subcatchments distributed across the south western counties of Cornwall, Devon, Somerset and Dorset were investigated. A suite of multivariate statistical techniques, including Direct (Redundancy Analysis) and Indirect (Principal Components Analysis) Ordination were used to explore catchment responses to four major storm events, selected from the wet autumn/winter of 2000-2001. A Geographic Information System. (GIS) incorporating the Hydrology of Soil Types (HOST) soil classification system and Land Cover Map 2000 satellite imagery data was developed to parameterise catchment physiographic variables and calculate the extent of 27 land use classes. Analysis of regional trends in erivironmental variables and two multivariate runoff datasets (R1 and R2) identified land use as the principal control of streamflow responses to extreme storm events. Land use, soil and geology parameters together explained 84% (R1) and 78% (R2) of the Variance in runoff for the same four storms. Grassland and improved grassland were consistent characteristics of catchments generating higher runoff volumes per unit area. Similarities in the hydrological behaviours of the Camel catchment and the De Lank subcatchment supported a dominant control on peak flows by runoff from grazed upland areas. A longer-term study of the River Camel catchment (1965-2000) revealed a 20% increase in the magnitude of the one in 25 year flow. Daily rainfall totals aggregated at monthly, seasonal and annual timescales and agricultural census data for the years 1969, 1979, 1988, 1997 and 2000 were examined to determine the influence of climate and land use changes on the enhanced streamflow response. Increases in the frequency and magnitude of peak flows were attributed to the cumulative impacts of a subtle, long-term rise in October rainfall totals, coupled with local urban development, the expansion of arable cultivation on highly connected slopes in the lower catchment and a rise in the intensity of grazing in the upper catchment At the field scale, characterisation of the textural, structural and ^hydraulic properties of soils subject to different land managements, including continuous cereal cultivation (CC), semi permanent pasture (SPP), permanent pasture (PP) and farm woodland (FW), identified a link between land use 'and the structural stability of the surface horizon. Marked differences in the percentage of water stable aggregates (WSA>2.8mm) between the topsoils of FW (66%) and.PP (71%), SPP (11%) and CC (6%) helped to explain differences in saturated hydraulic conductivity that were in the order FW>PP>SPP>CC, Laboratory rainfall simulations revealed slower wetting rates and higher average soil moisture percentages at near-saturation in FW and PP soil plots compared to SPP and CC soil plots that resulted from higher total porosities under FW and PP. Agricultural management systems are therefore capable of playing an important role in attenuating peak flow responses to storm events through considered land management which ameliorates or prevents soil structural deterioration and encourages the movement of water into storages within the hillslope. The adoption of specific measures, such as the introduction of buffer strips, widening of hedgerows or the introduction of forested areas to act as sinks, may serve to disconnect hydrofogical pathways from the main channel by providing a barrier to runoff, thereby reducing the upslope contributing area.

Airborne scanning laser altimetry (LiDAR) is an important new data source for many environmental applications, mapping topographic and surface object height to high vertical (±15-20cm) and horizontal (±30-100cm) accuracy over large areas, both time and cost effectively. These data offer improvements in 2D hydraulic flood models by providing floodplain bathymetry and vegetation height for the parameterisation of friction. Current parameterisations use one (temporally constant) value of friction for the floodplain, and one for the channel, with these values determined through a calibration procedure which limits the physical basis of the model and, hence, its applicability to different catchments and flood events. Primarily for this reason, a LiDAR data processing system is developed that segments a rural scene into water and three vegetation height classes. The vegetation and topographic heights in each class are calculated, and are accurate to ±14cm and ±17cm (respectively) for the class 'crops and grasses'. The vegetation heights are subsequently converted. using existing empirical equations, into friction coefficients that vary with the local flow depth and velocity. This friction parameterisation is implemented in the TELEMAC-2D model and a Hood that occurred on the river Severn in 1998 is simulated. When compared with a Synthetic Aperture Radar (SAR) image of the event, the model accurately predicts the inundation extent. The LiDAR vegetation segmentation is also used to drive a new mesh generator which decomposes the mesh around automatically identified vegetation features so t.hat. regions of abruptly changing boundary friction (and, as it. transpires, topographic gradient) are represented more explicitly in t.he model. Further model simulations demonstrate (i) an improvement in inundation extent prediction using the spatiotemporally variable friction parameterisation on the decomposed mesh, and (ii) variations in predicted velocities which may be observable using remote sensing. Indeed. velocity patterns are identified as important in model validation as the SAR data exhibit inadequacies.

This study presents an integrated approach to estimating the risk of flash floods in arid wadis. Flash floods impact seriously on the infrastructure and development of Egyptian desert towns, such as Marsa Alam, and their surrounding areas. The study developed a method using the example of the wadi El-Alam, which has an area of approximately 407 km2. The method has three stages. First, the construction of a Digital Elevation Model from which morphometric properties of the brain were calculated. In particular, representing the flat areas and the sharp junctions between badland slopes and the adjacent wadi system (which generally characterised arid landscapes), were among the major issues facing this study. Second, the derivation of land cover from satellite remote sensing data which were rigorously pre-processed with the aid of aerial photographs and fieldwork measurements. Third, the application of a hydrological model, incorporating the basin morphometry and land cover data, to transform rainfall into runoff and route the floodwater through the sub-catchments of the wadi. The hydrological model system (HMS) was run to simulate discharge at the main wadi outlet (where the town of Marsa Alam is located) and at each sub-basin intersecting the Idfu-Alam highway that runs through the wadi. From these results, sites vulnerable to flash flood and their risk classes were identified. The sites judged to have the highest flood hazard along the highway were exactly the sites damaged in the 1991 flash flood. Thus, as an outcome of this study, the first digital database that includes information on land cover, geomorphology and hydrology of wadi El-Alam was developed. This database can be used to highlight regions vulnerable to flood damage.

It is argued in this thesis that the climate signal shows more strongly in the runoff regime of the Gambia River Basin (GRB) than the signal from deforestation. Partial and multiple regression was used to partition the effects on runoff of rainfall characteristics and deforestation over the GRB since the turn of this century. The expected shorter, higher more rapidly responding wet season flood peaks which result from deforestation have not occurred in the GRB. Rather, peak floods have fluctuated since the beginning of the century, but showing a clear declining trend similar to the rainfall regime. The large size (z 7550 km') of the sub-catchments of the GRB inhibit synchronisation of the rapid runoff that is associated with deforestation. Furthermore, deforestation, as it occurs in the GRB, takes place piece-meal as small plots of land are cleared. The nature of clearance of vegetation is important; the vegetation cleared is either replaced with another type of vegetation, for example, groundnuts or millet, or is soon allowed to recover after a cropping phase. Surface and sub-surface hydrological processes within the GRB are therefore not subjected to the severe form of alteration that characterise massive and total clearance of vegetation schemes in urban development. However, deforestation has significantly affected low flowsthere are now longer periods of lower dry season flows, and these are ascribed to the diminishing recharge of ground water. By augmenting overland flow and reducing interception and infiltration, deforestation causes a reduction in ground water recharge, which is an important component of dry season flows. Both climate change and deforestation have worked in parallel to cause a fluctuating but declining flow regime of the Gambia River. This, in turn, affects both the agricultural potential and productivity of the GRB.

Urban drainage models are examined in terms of model purpose, type and usage. Model usage for the prediction of flooding behaviour in the UK adopts the design storm approach, which assumes that flood frequency equals rainfall frequency. This equation is shown to be affected by antecedent moisture conditions and rainfall spatial variability. Data collected from three small urban catchments (0.3 ha, 0.7 ha and 3.2 ha) in Southampton are used to calibrate and verify three urban drainage models: namely WASSP-SIM, WALLRUS-SIM and USGS-DR3M. A probabilistic approach to model verification is used and demonstrates that accepted bounds to model accuracy associated with the use of these models are difficult to justify if consideration is given to the possible errors in rainfall and discharge measurement. Comparison of the results from storms collected during 1985 on these catchments shows that the DR3M performs better than either of the other models in terms of a number of standard indices of model accuracy. Soil moisture observations collected over a two year period using the neutron probe method are then used to calibrate and test a number of simple soil moisture models including the soil moisture algorithm incorporated in the MORECS procedure and that included in the DR3M. These models are found to predict catchment average soil moisture conditions to within 20 mm, using local rainfall information and standard evapotranspiration curves as inputs. Using the MORECS soil moisture model it is shown that antecedent soil moisture conditions do have an impact on observed percentage runoff values, but that the relationship is confused by seasonal factors. The soil moisture model is then used together with WASSP-SIM to derive discharge-frequency curves using a continuous simulation approach. Comparisons with the design storm approach confirm the assumptions of the approach, although they do not prove that either approach produces the correct answers.

No description available.

Hydrological time series are often asymmetric in time, insomuch as rises are more rapid than recessions, as well as having highly skewed marginal distributions. A two-stage transformation is proposed for deseasonalised daily average flow series. Rises are stretched, and recessions are squashed until the series is symmetric over time. An autoregressive moving average (ARMA) model is then fitted to the natural logarithms of this new series The residuals from the ARMA model are represented by Weibull distributions. Seasonal flood risks, as daily average flows, are estimated by simulation. However, floods are often measured as peak flows rather than daily average flows, although both measures are relevant, and the use of growth factors to allow for this is demonstrated. The method is demonstrated with 24 years of daily flows from River Cherwell in the south of England, a 40-years record from the upper reaches of the Thames and 21-years record from the River Coquet in the north-east of England. Seasonal estimates of flood risk are given, and these can be conditioned on catchment wetness at the time of prediction. Comparisons with other methods which allow for time irreversibility are also made. One is ARMA models with exogenous input, in this case rainfall, which will, because of its intermittent nature, impact a natural time irreversibility to the streamflow series. A disadvantage of these models is that they require rainfall data in addition to the streamflow record. A second is the development of a class of shot noise models, which naturally generate highly time irreversibility series. This is the Neyman-Scott model. But, despite its attractive physical interpretation it is inevitably less flexible than the two stage transformation because it has fewer parameters. Although it was found to provide a good fit to daily data it is less convincing for the extremes. Overall the two stage transformation (TST) compared favourably with both models.