Floods are the most common natural disaster in the U.S. as reported by the Federal Emergency Management Administration (FEMA), and there is a need to provide advance warning to vulnerable communities on the potential risks of flooding after intense storms. The key drivers of urban hydrological research include climate change impacts and adaption, city resilience to hydrological extremes, and integration with emergency management and city planning disciplines. Significant advances in modeling techniques and computational resources have made real-time flood forecasting tools in urban and rural areas an achievable goal, but there is no universal method for flood modeling. Urban landscapes pose a challenge because of fine-scale features and heterogeneities in the landscape including streets, buildings, pipes, and impervious land cover. A nested regional-local modeling approach was used to evaluate its capabilities to provide useful and accurate flood related information to a small community in Iowa. The advantage of a nested approach is the ability to harness the computational efficiency of the regional model while providing reasonably accurate streamflow boundary conditions to the local model. The nested model incorporates the tools and products maintained at the Iowa Flood Center (IFC) including the streamflow bridge sensors, rain gauges, radar rainfall product, and statewide model. A one-way connection was made between the regional model of the upper Maquoketa Watershed (275 mi2) and the local model of the City of Manchester (5 mi2). The uncalibrated, nested model was validated using photos and streamflow records for flood events that occurred in July 2010 and September 2016. Multiple radar rainfall estimates were used as input to the model to better understand the impacts of the spatial and temporal resolution and variations of rainfall on streamflow predictions. A local storm event analysis was completed to determine the vulnerable areas of the stormwater network in eastern Manchester. The two main sources of flooding in Manchester are from the river and from local runoff. During extreme flood events caused by the river, the hydrologic impacts of the urban catchment are masked and the stormwater network system is overwhelmed. The coarse, regional model is limited in producing streamflow results for the small tributaries draining the eastern areas of Manchester. In the case of localized rainfall, a fine resolution model that takes into account the stormwater network and rainfall-runoff dynamics are crucial to capturing the hydrologic response of the urban area. Overall, the nested model showed skill in reproducing the hydrographs and the flood extents. Using an ensemble of rainfall input, the multiple model realizations envelope the observed streamflow indicating that the uncertainty of the rainfall is implicitly captured in the model results. The simulated streamflow at the outlet varies significantly depending on the spatial resolution of the rainfall but shows small sensitivity to the temporal resolution of the rainfall input. However, the local rainfall-runoff volumes vary significantly depending on the spatial and temporal resolution of the rainfall input. Recommendations are given to Manchester to highlight areas at risk to flooding. Recommendations are given to the IFC on the capabilities of the nested regional-local modeling approach along with suggestions for future work to incorporate urban areas into the statewide flood forecasting system.

Every dam must be capable of safely passing a predetermined flood magnitude. For high-hazard dams, it is a common practice to require that this flood be the Probable Maximum Flood (PMF). The determination of the PMF starts with the determination of the Probable Maximum Precipitation (PMP). We used the generalized estimates of the PMP as outlined in Hydrometeorological Report (HMR) 49. In this study we used the storm event model approach to convert the PMP into PMF. Different synthetic unit hydrograph (SUH) techniques were then used in the conversion process. The Soil Conservation Service (SCS) curve number method was used to estimate the losses, and the U.S. Bureau of Reclamation (USBR) lag time relationship was used as the basis for estimating the time parameters for the different (SUH) methods. The objectives of this study were 1) to evaluate the theoretical and empirical basis for the SUH methods that are commonly used for Utah PMF determinations; 2) to compare the PMF determinations for representative Utah watersheds based on alternative SUH methods using Geographic Information System (GIS) techniques, and to explore the effects of parameter uncertainty ; and 3) to make recommendations for the selection and use of SUH methods for Utah PMF determinations. An interactive PMF modeling system was formulated. The modeling system processes the different databases and estimates the parameters required for HEC 1 model input to produce the PMF hydrograph. Five SUHs were used, tbe SCS, Clark, Snyder, USBR, and the Corps of Engineers (COE) LA valley S-graph. Seven representative Utah high-hazard dam sites were selected and used in the evaluation of the five SUH techniques, focusing on their procedure, practice and applicability, and analytical and empirical evaluations. GIS procedures proved to be a very efficient and flexible means for obtaining rainfall-runoff model inputs. Deviation of the site-specific time-area curve from the default curve in HEC 1 leads to errors in the peak flow estimate. In the absence of suitable events for site-specific development of unit hydrographs, the USBR SUH technique is to be used, but careful consideration should be given to the appropriateness of the use of local storm K, values for Utah local storm PMFs.

Design flood estimation in small to medium sized ungauged catchments is frequently required in hydrologic analysis and design and is of notable economic significance. For this task Australian Rainfall and Runoff (ARR) 1987, the National Guideline for Design Flow Estimation, recommends the Probabilistic Rational Method (PRM) for general use in South- East Australia. However, there have been recent developments that indicated significant potential to provide more meaningful and accurate design flood estimation in small to medium sized ungauged catchments. These include the L moments based index flood method and a range of quantile regression techniques. This thesis focuses on the quantile regression techniques and compares two methods: ordinary least squares (OLS) and generalised least squares (GLS) based regression techniques. It also makes comparison with the currently recommended Probabilistic Rational Method. The OLS model is used by hydrologists to estimate the parameters of regional hydrological models. However, more recent studies have indicated that the parameter estimates are usually unstable and that the OLS procedure often violates the assumption of homoskedasticity. The GLS based regression procedure accounts for the varying sampling error, correlation between concurrent flows, correlations between the residuals and the fitted quantiles and model error in the regional model, thus one would expect more accurate flood quantile estimation by this method. This thesis uses data from 133 catchments in the state of Victoria to develop prediction equations involving readily obtainable catchment characteristics data. The GLS regression procedure is explored further by carrying out a 4-stage generalised least squares analysis where the development of the prediction equations is based on relating hydrological statistics such as mean flows, standard deviations, skewness and flow quantiles to catchment characteristics. This study also presents the validation of the two techniques by carrying out a split-sample validation on a set of independent test catchments. The PRM is also tested by deriving an updated PRM technique with the new data set and carrying out a split sample validation on the test catchments. The results show that GLS based regression provides more accurate design flood estimates than the OLS regression procedure and the PRM. Based on the average variance of prediction, standard error of estimate, traditional statistics and new statistics, rankings and the median relative error values, the GLS method provided more accurate flood frequency estimates especially for the smaller catchments in the range of 1-300 km2. The predictive ability of the GLS model is also evident in the regression coefficient values when comparing with the OLS method. However, the performance of the PRM method, particularly for the larger catchments appears to be satisfactory as well. / Master of Engineering (Honours)

<p>A big responsibility lies in the hand of local authorities to exercise measures in preventing fatalities and damages during flood occurrences. However, the problem is how flooding can be prevented if nobody knows when and where it will be occurring, and how much water is expected. Therefore, the utilisation of flood models in such studies can be helpful in simulating what is anticipated to occur.</p><p> </p><p>In this study, the HEC-RAS steady flow model was used in calibrating different flood events in Testeboån river, which is situated in the municipality of Gävle in Sweden. The purpose is to provide inundation maps that show the water surface profiles for the various flood events that can help authorities in planning within the area. Moreover, the study would try to address certain issues, which concern one-dimensional models like HEC-RAS in terms of the effects of topographic data and the parameters used for friction coefficient.</p><p> </p><p>Various flood maps were produced to visualise the extents of the floods. In Oppala and Norra Åbyggeby, the big water extents for both the 100-year and the highest probable floods were visible in the forested areas and grasslands, although a few houses were within the predicted flooded areas. In Södra Åbyggeby, Varva, Forsby, and in the northern parts of Strömsbro and Stigslund, the majority of the residential places were not inundated during the 100-year flood calibration, but became flooded during the maximum probable flood. The southern portions of Strömsbro and Stigslund had lesser flood extents and houses were situated within the boundaries of the highest flood. In Näringen, there were also some areas close to the estuary that were flooded for both events.</p><p> </p><p>With the other calibrations performed, two factors that greatly affect the flood extents in the floodplain, particularly in flatter areas were topographic data and the parameters used as friction coefficient.  The use of high resolution topographic data was important in improving the performance of the software. Nevertheless, it must be emphasised that in areas characterised by gentler slopes that bounded the channel and the floodplain, data completeness became significant whereby both ground data and bathymetric points must be present to avoid overestimation of the inundation extent. The water extents also varied with the use of the various Manning’s <em>n</em> for the overbanks, with the bigger value showing greater water extents. Else, in areas with steeper slopes and where the water was confined to the banks, the effect was minimal.</p><p> </p><p>Despite these shortcomings of one-dimensional models, HEC-RAS provided good inundation extents that were comparable to the actual extent of the 1977 flooding.</p><p> </p><p>Modelling real floods has its own difficulties due to the unpredictability of real-life flood behaviours, and more especially, there are time dependent factors that are involved.  Although calibrating a flood event will not exactly determine what is to arise as they might either under- or overestimate such flooding occurrences, still, they give a standpoint of what is more or less to anticipate, and from this,  planning measures can be undertaken.</p>

Thesis (Ph.D.)--University of Wisconsin--Madison, 2006. / Includes bibliographical references (p. 265-291). Also available on the Internet.

Floods can result in enormous causalities and huge economic losses in urban and rural regions. In recent years, while assessing the damage, risk analysis and assessment has become an important tool in addressing uncertainty in flood hazards. The lack of knowledge about the water extend, water depth, water velocity and potential damage in case of flood increase the vulnerability of the people to disasters in the flood region. Especially this information is valuable for city planners and decision makers. In case of new settlement area selection, correct decision can be taken by the help of this information. This type of information can be taken from hydraulic models as 1D or 2D. On the other hand, two dimensional (2D) hydraulic modeling becomes a need with increasing trends of very high speed computers and models instead of one dimensional (1D) ones. The ability of solving complex structures within few minutes enhances the use of 2D modeling with the integration of wave motion. In addressing the uncertainty, GIS becomes an important tool in risk assessment by integrating the flood depth, extend and vulnerability issues for definition of the quantitative risk. In this study, 1D and 2D hydraulic modeling is applied and combined with the quantitative vulnerability factors in Dalaman Plain-Turkey. Results show that the area is vulnerable to flood and high monetary damages can be seen in case a flood in the region.

Översvämningar förekommer regelbundet på stora delar av jordklotet. Utgående från de klimatförändringar vi ser idag med, bland annat, kraftigare och intensivare nederbörd är det troligt att det blir fler och större översvämningar framöver. Områden som inte tidigare varit översvämmade kan komma att bli översvämmande.   Varje land eller område med regelbundna översvämningar har någon form av varningssystem. Översikter över olika typer av system för att varna för översvämningar är svåra att finna.   Syftet med denna uppsats är att: Undersöka viktiga principer för effektiva varningssystem för översvämningar som orsakas av kraftig nederbörd och/eller vårflöden. Skapa en systematisk och strukturerad sammanställning över nuvarande system för att varna för översvämningar orsakade av kraftig nederbörd och/eller vårflöden. Ge praktiska exempel på några svenska kommuners system för att varna för översvämningar.   Litteratur- och internetsökningar kring varningssystem för översvämningar samt intervjuer med ”översvämningsansvariga” i Ödeshög och Vetlanda kommuner utgör grunden för informationen i denna uppsats.   Exempel på olika typer av varningssystem för översvämningar presenteras. Saker att tänka på vid val av, och vid drift och underhåll av, automatiska varningssystem belyses.   En indelning (Grust, 2006) av varningssystem i manuella, enkla automatiska och sofistikerade automatiska utökas till en matris med i matrisens andra dimension lokala, avrinningsområdes, nationella, multinationella och globala varningssystem. De i uppsatsen beskrivna exemplen på varningssystem positioneras in i matrisen.   Varningssystem för översvämningar i två svenska kommuner, Ödeshög och Vetlanda, beskrivs och likheter och skillnader mellan kommunernas system presenteras. / Flooding occurs all over the world on a regular basis. Based on the climate change ongoing today with heavier and more intense precipitation we can expect more and larger floods than we have seen before. Areas which have not been flooded earlier may now become flooded.   Every country or area which is flooded on a regular basis has some kind of warning system. Overviews of various types of flood warning systems are difficult to find.   The purpose of this thesis is to: Investigate important principles of effective flood warning systems caused by heavy precipitation and/or spring flood. Create a structured and systematical overview of current flood warning system for floods caused by heavy precipitation and/or spring flood. Describe two local Swedish community´s flood warning systems.   Literature and Internet search covering flood warning systems combined with interviews on site at Ödeshög and Vetlanda forms the basis for the information in this thesis.   Examples of various types of flood warning systems are presented. Issues regarding the selection of, and the operation and maintenance of, automatic warning systems are described.   A grouping (Grust, 2006) of warning systems in manual, simple automatic and sophisticated automatic is expanded to a matrix with the second dimension of the matrix being local, drainage area, national, multinational and global warning systems. The examples of warning systems covered in this thesis are placed in the matrix.   The similarities and differences between the flood warning systems in two local Swedish communities Ödeshög and Vetlanda are described.

From the beginning of 1980s, Hsichin has become a dreaming heaven of emigrants who move to the urban area of Taipei because the fever of housing market and the policy of government. As the high density development of slope land, the runoff capacity will be increased. On the other hand, river land will be decreased by city growth. Impacting by the above two reason, Hsichin is called as the flown city by the mass media. The goal of this research is discussing the reaction between the city development of Hsichin, the flood event, and the flood prevention activities. This research uses ¡§System Dynamics¡¨ as a research method, simulating with ¡§iThink¡¨ computer software, building a model of Hsichin city growth and flood event, using this model to precede the policy test to find out which policy will be the economic benefit. The research conclusion is: 1. The aspect of engineering activities: We adopt three policies including Yuansantze activity, dike building activity, and dredging activity. Under the same cost, we¡¦ll find out Yuansantze activity is the most benefit one. When mixing all policies, the most benefit one is putting the fund as 25% in dike building and 75% in dredging. These activities are focused on lowering flood damage directly. In the short term, the outcome is very obvious. However, in the long term, combining city growth management will be the only solution to solve problems of not only city overgrowth but also land development disequilibrium. 2. The aspect of city growth management: We adopt three policies including slope land conservation, converting land to flood storage area, and limiting of population and business. By this research, we¡¦ll find out limiting of population and business is the most efficient method focused on controlling population scale and lowering land development. This research provides an alternative way which is combining limiting high density development and slope land conservation to solve the situation of land development disequilibrium. In my opinion, the authority government cannot take engineering activities as the only way to control flood damage, however, the government should combine two aspects of controlling flood activities to solve Hsichin flood problem.

A big responsibility lies in the hand of local authorities to exercise measures in preventing fatalities and damages during flood occurrences. However, the problem is how flooding can be prevented if nobody knows when and where it will be occurring, and how much water is expected. Therefore, the utilisation of flood models in such studies can be helpful in simulating what is anticipated to occur.   In this study, the HEC-RAS steady flow model was used in calibrating different flood events in Testeboån river, which is situated in the municipality of Gävle in Sweden. The purpose is to provide inundation maps that show the water surface profiles for the various flood events that can help authorities in planning within the area. Moreover, the study would try to address certain issues, which concern one-dimensional models like HEC-RAS in terms of the effects of topographic data and the parameters used for friction coefficient.   Various flood maps were produced to visualise the extents of the floods. In Oppala and Norra Åbyggeby, the big water extents for both the 100-year and the highest probable floods were visible in the forested areas and grasslands, although a few houses were within the predicted flooded areas. In Södra Åbyggeby, Varva, Forsby, and in the northern parts of Strömsbro and Stigslund, the majority of the residential places were not inundated during the 100-year flood calibration, but became flooded during the maximum probable flood. The southern portions of Strömsbro and Stigslund had lesser flood extents and houses were situated within the boundaries of the highest flood. In Näringen, there were also some areas close to the estuary that were flooded for both events.   With the other calibrations performed, two factors that greatly affect the flood extents in the floodplain, particularly in flatter areas were topographic data and the parameters used as friction coefficient.  The use of high resolution topographic data was important in improving the performance of the software. Nevertheless, it must be emphasised that in areas characterised by gentler slopes that bounded the channel and the floodplain, data completeness became significant whereby both ground data and bathymetric points must be present to avoid overestimation of the inundation extent. The water extents also varied with the use of the various Manning’s n for the overbanks, with the bigger value showing greater water extents. Else, in areas with steeper slopes and where the water was confined to the banks, the effect was minimal.   Despite these shortcomings of one-dimensional models, HEC-RAS provided good inundation extents that were comparable to the actual extent of the 1977 flooding.   Modelling real floods has its own difficulties due to the unpredictability of real-life flood behaviours, and more especially, there are time dependent factors that are involved.  Although calibrating a flood event will not exactly determine what is to arise as they might either under- or overestimate such flooding occurrences, still, they give a standpoint of what is more or less to anticipate, and from this,  planning measures can be undertaken.

published_or_final_version / Civil Engineering / Master / Master of Philosophy