Thesis (MEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: The Francou and Rodier (1967) empirical approach uses the original concept of envelope curves for the definition of the regional maximum flood (RMF). Kovacs (1980) adopted the Francou and Rodier empirical flood calculation method and applied it to 355 catchments in South Africa. He revised his study in 1988 to also include the southern portions of the Southern Africa subcontinent. No method other than the Francou and Rodier empirical flood approach in the reviewed literature was found to be suitable for the purpose of this study. Therefore the Francou and Rodier empirical approach, as applied by Kovacs in 1988, was reapplied and used in this study to update the RMF for Lesotho. Maximum recorded flood peaks were derived from annual maximum time series and an up to date catalogue of flood peaks for 29 catchments was compiled for Lesotho. The maximum recorded flood peaks were then plotted on the logarithmic scale against their corresponding catchment areas. There are 3 major river systems that divide Lesotho into hydrologically homogenous basins. Envelope curves were drawn on the upper bound of the cloud of plotted points for these 3 river basins. These envelope curves represent the maximum flood peaks that can reasonably be expected to occur within the respective river basins in Lesotho. / AFRIKAANSE OPSOMMING: Francou en Rodier (1967) se empiriese benadering maak gebruik van die oorspronklike konsep van boonste limiet kurwes vir die definisie van die streeks maksimum vloed (SMV). Kovacs (1980) het die Francou en Rodier empiriese vloed berekening metode toegepas op 355 opvanggebiede in Suid-Afrika. Hy hersien sy studie in 1988 om ook die suidelike gedeeltes van die Suider-Afrikaanse subkontinent in te sluit. Geen ander metode as die Francou en Rodier empiriese vloed benadering is in die literatuur gevind wat as geskik aanvaar kan word vir die doel van hierdie studie nie. Daarom is die Francou en Rodier empiriese benadering, soos toegepas deur Kovacs in 1988, weer in hierdie studie toegepas en gebruik om die SMV metode vir Lesotho op te dateer. Maksimum aangetekende vloedpieke is verkry vanuit jaarlikse maksimum tyd-reekse en ʼn opgedateerde katalogus van vloedpieke vir 29 opvanggebiede saamgestel vir Lesotho. Die maksimum aangetekende vloedpieke is grafies aangetoon op logaritmiese skaal teenoor hul opvanggebiede. Daar is 3 groot rivierstelsels wat Lesotho in hidrologiese homogene gebiede verdeel. Boonste limiet kurwes is opgestel om die boonste grens van die gestipte punte vir hierdie 3 gebiede aan te toon. Hierdie krommes verteenwoordig die maksimum vloedpieke wat redelikerwys verwag kan word om binne die onderskeie rivierstelsels in Lesotho voor te kan kom.

Throughout history floods have been part of the human experience. They are recurring phenomena that form a necessary and enduring feature of all river basin and lowland coastal systems. In an average year, they benefit millions of people who depend on them. In the more developed countries, major floods can be the largest cause of economic losses from natural disasters, and are also a major cause of disaster-related deaths in the less developed countries. Flood disaster mitigation research was conducted to determine how remotely sensed data can effectively be used to produce accurate flood plain maps (FPMs), and to identify/quantify the sources of error associated with such data. Differences were analyzed between flood maps produced by an automated remote sensing analysis tailored to the available satellite remote sensing datasets (rFPM), the 100-year flooded areas "predicted" by the Flood Insurance Rate Maps, and FPMs based on DEM and hydrological data (aFPM). Landuse/landcover was also examined to determine its influence on rFPM errors. These errors were identified and the results were integrated in a GIS to minimize landuse / landcover effects. Two substantial flood events were analyzed. These events were selected because of their similar characteristics (i.e., the existence of FIRM or Q3 data; flood data which included flood peaks, rating curves, and flood profiles; and DEM and remote sensing imagery.) Automatic feature extraction was determined to be an important component for successful flood analysis. A process network, in conjunction with domain specific information, was used to map raw remotely sensed data onto a representation that is more compatible with a GIS data model. From a practical point of view, rFPM provides a way to automatically match existing data models to the type of remote sensing data available for each event under investigation. Overall, results showed how remote sensing could contribute to the complex problem of flood management by providing an efficient way to revise the National Flood Insurance Program maps.

A previsão hidrológica possibilita a identificação antecipada de eventos de cheia potencialmente causadores de inundação, o que é de grande importância para a atuação de entidades como a Defesa Civil. Quando se deseja estender a antecedência no tempo em relação a estes eventos, principalmente nos casos onde a bacia de interesse é relativamente rápida, torna-se necessária a incorporação de previsões quantitativas de precipitação (QPF) na modelagem hidrológica, as quais podem ser obtidas a partir de modelos numéricos de previsão do tempo. Entretanto, a falta de acurácia atribuída a estas previsões de chuva, dadas de forma determinística, vem promovendo sua substituição por sistemas de previsão meteorológica por conjunto (EPS - Ensemble Prediction Systems), cuja finalidade é a geração de possíveis estados futuros da atmosfera para considerar as incertezas associadas ao seu estado inicial e às deficiências na representação física dos modelos de previsão do tempo. Neste contexto, o presente estudo teve por objetivo avaliar uma metodologia de previsão de cheias por conjunto na bacia do Taquari-Antas/RS até a cidade de Encantado (19.000 km²), localizada na região Sul do Brasil. Para tanto, foi utilizado o modelo hidrológico MGB-IPH acoplado a diferentes sistemas de previsão, sendo eles: (i) EPS Regional ETA, de curto prazo (até 72 horas) com 5 membros de diferentes parametrizações; (ii) EPS Global ECMWF de médio prazo (até 10 dias) com 50 membros de condições iniciais perturbadas, incluindo perturbação estocástica nos parâmetros de ajuste do modelo e; (iii) Previsão Determinística do Modelo Regional ETA (até 7 dias). A avaliação das previsões consistiu em dois hindcastings distintos, envolvendo uma análise visual de eventos singulares ocorridos em 06/06/2014 e 21/07/2011 além de uma análise estatística no período de Mar/2014 - Nov/2014. Durante a análise visual foi possível identificar, a partir de antecedências de 5 a 6 dias, uma persistência na previsão dada pelo crescente número de membros do conjunto de médio prazo (ECMWF) com superação dos limiares de referência, na medida em que se aproximavam os eventos de cheia. Apesar da grande incerteza na magnitude das previsões hidrológicas para o conjunto de curto prazo, a vazão máxima foi relativamente bem prevista por pelo menos 1 membro em quase todas as antecedências, enquanto que a previsão do timing dos eventos foi considerada de boa confiabilidade. Durante a avaliação estatística foi possível notar uma falta de espalhamento nos conjuntos, com tendência de subestimativa de acordo com o aumento da antecedência. Em uma comparação com previsões determinísticas, as previsões por conjunto demonstraram maior acurácia principalmente até 72 horas de antecedência, com destaque para a maior probabilidade de detecção dos limiares de referência e manutenção de falso alarme a níveis reduzidos. Além disso, verificou-se também que a agregação de previsões efetuadas em tempo anterior naquelas atuais acarreta em ligeira ampliação do espalhamento do conjunto e maiores probabilidades de detecção dos limiares de alerta para os membros mais elevados, apesar da redução no desempenho em termos de acurácia e viés. De forma geral, as previsões por conjunto apresentam potencial para servir como uma informação complementar em sistemas de alerta contra cheias, possibilitando uma melhor preparação dos agentes envolvidos durante a ocorrência destes eventos. / Hydrological forecasting plays an important role for issuing flood warnings, allowing for anticipation and better preparation of authorities at the occurrence of such events. In order to extend lead time in a flood forecast, especially when the catchment response time is relatively fast, it may be useful to couple a hydrological model to quantitative precipitation forecasts (QPF), usually obtained directly from numerical weather prediction (NWP). However, deterministic (i.e. single) QPF are usually referred to many errors and lack of accuracy, mainly caused by uncertainties on initial state of the atmosphere and on physical representation of weather forecasting models. To address these shortcomings, it becomes necessary to take into account the uncertainties associated to rain forecasts, which can be represented by Ensemble Prediction Systems (EPS). The purpose of such systems is to provide different trajectories of the atmosphere by perturbations on its initial condition and on parameterization schemes of the models, generating an ensemble of forecasts that can be used as input to hydrological modelling (HEPS). In this context, the present study aimed to assess a methodology of ensemble flood forecasting on Taquari-Antas basin up to the city of Encantado/RS (19.000 km²), located in southern Brazil. Therefore, the hydrological model MGB-IPH was coupled to different forecasting systems: (i) Short Range EPS ETA (up to 72 hours), a regional model with 5 members of different parameterization schemes; (ii) Medium Range EPS ECMWF (up to 10 days), a global model with 50 members of perturbed initial conditions and stochastic perturbation in the model parameters; (iii) Deterministic ETA Model (up to 7 days). The forecasts were evaluated by two different hindcastings, which includes a visual assessment of singular events occurred on 2011 and 2014 and a statistical analysis for the period between Mar/2014 and Nov/2014. It was possible to identify a forecast persistence on medium-range for the selected events, by the increasing number of members exceeding the reference thresholds from lead times up to 5 - 6 days. On the short range, although large uncertainties in the magnitude of hydrological forecasts were found, the peak discharge was well forecasted - at least for a single member - in nearly all lead times, whereas the prediction of the peak timing was considered reliable. Regarding to statistical evaluation, an inadequate spread in the ensemble was observed from short- to medium-range, with a tendency of underestimation for longer lead times. In a comparison with deterministic forecasts, the ensemble forecasts showed higher accuracy especially up to 72 hours in advance, including highlights on greater probability of detection (POD) above the reference thresholds even with low false alarm rates. It also was found that the a combination of previous forecasts on the recent ones leads to a slight increase of ensemble spread and POD for higher members, despite the performance reduction in terms of accuracy and bias. In summary, the hydrological ensemble forecasts demonstrated a good potential to serve as an additional information within a Flood Alert System.

"May 2002" Includes list of papers published during this study Errata slip inserted inside back cover of v. 1 Includes bibliographical references (leaves 331-357) V. 1. [Text} -- v. 2. Appendices Develops an alternative design flood estimation methodology. Establishing a relationship between catchment characteristics and the rainfall excess frequency duration proportions enables the definition of these proportions for generic catchment types, increasing the potential for translation to catchments with limited data but similar hydrographic properties, thereby improving design process.

"May 2002" / Includes list of papers published during this study / Errata slip inserted inside back cover of v. 1 / Includes bibliographical references (leaves 331-357) / 2 v. : ill. (some col.), col. maps ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Develops an alternative design flood estimation methodology. Establishing a relationship between catchment characteristics and the rainfall excess frequency duration proportions enables the definition of these proportions for generic catchment types, increasing the potential for translation to catchments with limited data but similar hydrographic properties, thereby improving design process. / Thesis (Ph.D.)--University of Adelaide, Dept. of Civil and Environmental Engineering, 2002

One of the major problems for the implementation of water resources planning and management in arid and semi-arid environments is the scarcity of hydrological data and, consequently, research studies. In this thesis, the hydrology of dryland river systems was analyzed and a semi-distributed hydrological model and a forecasting approach were developed for flow transmission processes in river-systems with a focus on semi-arid conditions. Three different sources of hydrological data (streamflow series, groundwater level series and multi-temporal satellite data) were combined in order to analyze the channel transmission losses of a large reach of the Jaguaribe River in NE Brazil. A perceptual model of this reach was derived suggesting that the application of models, which were developed for sub-humid and temperate regions, may be more suitable for this reach than classical models, which were developed for arid and semi-arid regions. Summarily, it was shown that this river reach is hydraulically connected with groundwater and shifts from being a losing river at the dry and beginning of rainy seasons to become a losing/gaining (mostly losing) river at the middle and end of rainy seasons. A new semi-distributed channel transmission losses model was developed, which was based primarily on the capability of simulation in very different dryland environments and flexible model structures for testing hypotheses on the dominant hydrological processes of rivers. This model was successfully tested in a large reach of the Jaguaribe River in NE Brazil and a small stream in the Walnut Gulch Experimental Watershed in the SW USA. Hypotheses on the dominant processes of the channel transmission losses (different model structures) in the Jaguaribe river were evaluated, showing that both lateral (stream-)aquifer water fluxes and ground-water flow in the underlying alluvium parallel to the river course are necessary to predict streamflow and channel transmission losses, the former process being more relevant than the latter. This procedure not only reduced model structure uncertainties, but also reported modelling failures rejecting model structure hypotheses, namely streamflow without river-aquifer interaction and stream-aquifer flow without groundwater flow parallel to the river course. The application of the model to different dryland environments enabled learning about the model itself from differences in channel reach responses. For example, the parameters related to the unsaturated part of the model, which were active for the small reach in the USA, presented a much greater variation in the sensitivity coefficients than those which drove the saturated part of the model, which were active for the large reach in Brazil. Moreover, a nonparametric approach, which dealt with both deterministic evolution and inherent fluctuations in river discharge data, was developed based on a qualitative dynamical system-based criterion, which involved a learning process about the structure of the time series, instead of a fitting procedure only. This approach, which was based only on the discharge time series itself, was applied to a headwater catchment in Germany, in which runoff are induced by either convective rainfall during the summer or snow melt in the spring. The application showed the following important features: • the differences between runoff measurements were more suitable than the actual runoff measurements when using regression models; • the catchment runoff system shifted from being a possible dynamical system contaminated with noise to a linear random process when the interval time of the discharge time series increased; • and runoff underestimation can be expected for rising limbs and overestimation for falling limbs. This nonparametric approach was compared with a distributed hydrological model designed for real-time flood forecasting, with both presenting similar results on average. Finally, a benchmark for hydrological research using semi-distributed modelling was proposed, based on the aforementioned analysis, modelling and forecasting of flow transmission processes. The aim of this benchmark was not to describe a blue-print for hydrological modelling design, but rather to propose a scientific method to improve hydrological knowledge using semi-distributed hydrological modelling. Following the application of the proposed benchmark to a case study, the actual state of its hydrological knowledge and its predictive uncertainty can be determined, primarily through rejected hypotheses on the dominant hydrological processes and differences in catchment/variables responses. / Die Bewirtschaftung von Wasserressourcen in ariden und semiariden Landschaften ist mit einer Reihe besonderer Probleme konfrontiert. Eines der größten Probleme für die Maßnahmenplanung und für das operationelle Management ist der Mangel an hydrologischen Daten und damit zusammenhängend auch die relativ kleine Zahl wissenschaftlicher Arbeiten zu dieser Thematik. In dieser Arbeit wurden 1) die grundlegenden hydrologischen Bedingungen von Trockenflusssystemen analysiert, 2) ein Modellsystem für Flüsse unter semiariden Bedingungen, und 3) ein nichtparametrisches Vorhersage-verfahren für Abflussvorgänge in Flüssen entwickelt. Der Wasserverlust in einem großen Abschnitt des Jaguaribe Flusses im nordöstlichen Brasilien wurde auf Basis von Daten zu Abflussraten, Grundwasserflurabstände und mit Hilfe multitemporaler Satellitendaten analysiert. Dafür wurde zuerst ein konzeptionelles hydrologisches Modell über die Mechanismen der Transferverluste in diesem Abschnitt des Trockenflusses erstellt. Dabei ergab sich, dass der Flussabschnitt mit dem Grundwasser hydraulisch verbunden ist. Der Flussabschnitt weist in der Trockenenzeit und am Anfang der Regenzeit nur Wasserverlust (Sickerung) zum Grundwasser auf. Im Laufe der Regenzeit findet auch ein gegenseitiger Austausch vom Grundwasser mit dem Flusswasser statt. Aufgrund dieser hydraulischen Kopplung zwischen Flusswasser und Grundwasser sind für diesen Flussabschnitt hydrologische Modellansätze anzuwenden, die generell für gekoppelte Fluss-Grundwassersysteme, v.a. in feuchtgemäßigten Klimaten, entwickelt wurden. Es wurde ein neuartiges hydrologisches Simulationsmodell für Transferverluste in Trockenflüssen entwickelt. Dieses Modell ist für unterschiedliche aride und semiaride Landschaften anwendbar und hat eine flexible Modellstruktur, wodurch unterschiedliche Hypothesen zur Relevanz einzelner hydrologische Prozesse getestet werden können. Es wurde für den zuvor genannten großen Abschnitt des Jaguaribe Flusses im nordöstlichen Brasilien und für einen kleinen Flussabschnitt im „Walnut Gulch Experimental Watershed“ (WGEW) in Arizona, Südwest-USA, angewendet. Für die eine prozess-orientierte Simulation von Abflussbedingungen und Transferverlusten im Einzugsgebiet des Jaguaribe hat sich gezeigt, dass die am besten geeignete Modellstruktur sowohl den Austausch zwischen Flusswasser und Grundwasser (senkrecht zur Fließrichtung des Flusses) als auch die parallel zum Fluss verlaufende Grundwasserströmung enthält. Die Simulationsexperimente mit unterschiedlichen Modellstrukturen („Hypothesentest“) reduzierte nicht nur die Modellstrukturunsicherheit, sondern quantifizierte auch die Qualität der Modellergebnisse bei folgenden Varianten der Modellstruktur: a) Abflluss im Fluss ohne Interaktion mit dem Grundwasser (keine Transferverluste) und b) Interaktion zwischen Fluss und Grundwasser ohne parallelen Grundwasserstrom zum Flussstrom. Durch die Anwendung auf die beiden unterschiedlichen Trockenflusssysteme wurden neue Erkenntnisse über die Sensitivität des Modells unter verschiedenen Bedingungen erworben. Beispielsweise waren die Parameter der ungesättigten Zone, die von hoher Relevanz für den kleinen Flussabschnitt im WGEW waren, viel sensitiver als die Parameter der gesättigten Zone, die besonders relevant für den Jaguaribe Flussabschnitt in Brasilien waren. Die Ursache für diese sehr unterschiedliche Sensitivität liegt darin, dass beim WGEW das Flusswasser nur mit der ungesättigten Zone in Kontakt steht, da sich in diesem Gebiet, welche im Vergleich zur Jaguaribe-Region noch deutlich trockener ist, kein Grund-wasserleiter bildet. Letztlich wurde ein nicht-parametrisches Verfahren, zur Simulation der deterministischen Evolution und stochastischen Fluktuation der Abflussdynamik entwickelt. Im Unterschied zu prozessbasiertem Modellsystemen basiert dieses Verfahren nicht auf Modellkalibrierung sondern auf einem Lernprozess, basierend auf Zeitreihendaten. Als Anwendungsbeispiel wurde ein mesoskaliges Einzugsgebiet im Erzgebirge, NO-Deutschland gewählt, in dem starke Abflussereignisse entweder durch konvektive Niederschlagsereignisse oder durch Schneeschmelze generiert werden. Die folgenden wichtigsten Ergebnisse wurden erzielt: • Regressionsmodellansätze basierend auf den zeitlichen Änderungen der Abflüsse liefern bessere Ergebnisse gegenüber Ansätzen basierend auf direkten Abflussdaten; • mit zunehmendem Vorhersagehorizont wandelt sich das hydrologische System von einem mit Zufallsanteilen verrauschten dynamischen System zu einem linearen probabilistischen Zufallsprozess; • Bei zunehmendem Abfluss (ansteigenden Ganglinie) erfolgt meist eine Abflussunterschätzung, bei abnehmendem Abfluss (fallende Ganglinie) erfolgt meist eine Abflussüberschätzung. Dieses nichtparametrische Verfahren ergibt im Vergleich mit einem prozess-orientierten und flächenverteilten hydrologischen Hochwasservorhersagemodell bis zu einem Vorhersagezeitraum von 3 Stunden Ergebnisse von vergleichbar guter Qualität. Letztendlich wurde ein Vorgehen bzgl. künftiger Forschungen zu hydrologischer Modellierung vorgeschlagen. Das Ziel dabei war ein wissenschaftliches Verfahren zur Verbesserung des hydrologischen Wissens über ein Einzugsgebiet. Diese Verfahren basiert auf einem Hypothesentest zu den relevanten hydrologischen Prozessen und der Untersuchung der Sensitivitäten der hydrologischen Variablen bei unterschiedlichen Einzugsgebieten.

A detailed and comprehensive description of the state of the art in the field of flood forecasting opens this work. Advantages and shortcomings of currently available methods are identified and discussed. Amongst others, one important aspect considers the most exigent weak point of today’s forecasting systems: The representation of all the fundamentally different event specific patterns of flood formation with one single set of model parameters. The study exemplarily proposes an alternative for overcoming this restriction by taking into account the different process characteristics of flood events via a dynamic parameterisation strategy. Other fundamental shortcomings in current approaches especially restrict the potential for real time flash flood forecasting, namely the considerable computational requirements together with the rather cumbersome operation of reliable physically based hydrologic models. The new PAI-OFF methodology (Process Modelling and Artificial Intelligence for Online Flood Forecasting) considers these problems and offers a way out of the general dilemma. It combines the reliability and predictive power of physically based, hydrologic models with the operational advantages of artificial intelligence. These operational advantages feature extremely low computation times, absolute robustness and straightforward operation. Such qualities easily allow for predicting flash floods in small catchments taking into account precipitation forecasts, whilst extremely basic computational requirements open the way for online Monte Carlo analysis of the forecast uncertainty. The study encompasses a detailed analysis of hydrological modeling and a problem specific artificial intelligence approach in the form of artificial neural networks, which build the PAI-OFF methodology. Herein, the synthesis of process modelling and artificial neural networks is achieved by a special training procedure. It optimizes the network according to the patterns of possible catchment reaction to rainstorms. This information is provided by means of a physically based catchment model, thus freeing the artificial neural network from its constriction to the range of observed data – the classical reason for unsatisfactory predictive power of netbased approaches. Instead, the PAI-OFF-net learns to portray the dominant process controls of flood formation in the considered catchment, allowing for a reliable predictive performance. The work ends with an exemplary forecasting of the 2002 flood in a 1700 km² East German watershed.

South Africa experiences some of the lowest water resource system yields in the world as a result of the high regional variability of rainfall and runoff. Population growth and economic development are placing increasing demands on the nation's scarce water resources. These factors, combined with some of the objectives of the new National Water Act (1998), are highlighting the need for efficient management of South Africa's water resources. In South Africa's National Water Act (1998) it is stated that its purpose is to ensure that the nation's water resources are protected, used, conserved, managed and controlled in a way, which takes into account, inter alia, i. promoting the efficient, sustainable and beneficial use of water in the public interest, and ii. managing floods and droughts. Efficient and sustainable water resource and risk management can be aided by the application of runoff forecasting. Forecasting thus fits into the ambit of the National Water Act and, therefore, there is a need for its operational application to be investigated. In this document an attempt is made to test the following hypotheses: Hypothesis 1: Reliable and skilful hydrological forecasts have the ability to prevent loss of life, spare considerable hardship and save affected industries and commerce millions of Rands annually if applied operationally within the context of water resources and risk management. Hypothesis 2: Long to medium term rainfall forecasts can be made with a degree of confidence, and these rainfall forecasts can be converted into runoff forecasts which, when applied within the framework of water resources and risk management, are more useful to water resource managers and users than rainfall forecasts by themselves. The validity of Hypothesis 1 is investigated by means of a literature review. South Africa's high climate variability and associated high levels of uncertainty as well as its current and future water resources situation are reviewed in order to highlight the importance of runoff forecasting in South Africa. Hypothesis 1 is further examined by reviewing the concepts of hazards and risk with a focus on the role of effective risk management in preventing human, financial and infrastructural losses. A runoff forecasting technique using an indirect methodology, whereby rainfall forecasts are translated into runoff forecasts, was developed in order to test Hypothesis 2. The techniques developed are applied using probabilistic regional rainfall forecasts supplied by the South African Weather Service for 30 day periods and categorical regional forecasts for one, three and four month periods for I regions making up the study area of South Africa, Lesotho and Swaziland. These forecasts where downscaled spatially for application to the 1946 Quaternary Catchments making up the study area and temporally to give daily rainfall forecast values. Different runoff forecasting time spans produced varying levels of forecast accuracy and skill, with the three month forecasts producing the worst results, followed by the four month forecasts. The 30 day and one month forecasts for the most part produced better results than the more extended forecast periods. In the study it was found that hydrological forecast accuracy results seem to be inversely correlated to the amount of rainfall received in a region, i.e. the wetter the region the less accurate the runoff forecasts. This trend is reflected in both temporal and spatial patterns where it would seem that variations in the antecedent moisture conditions in wetter areas and wetter periods contribute to the overall variability, rendering forecasts less accurate. In general, the runoff forecasts improve with corresponding improvements in the rainfall forecast accuracy. There are, however, runoff forecast periods and certain regions that produce poor runoff forecast results even with improved rainfall forecasts. This would suggest that even perfect rainfall forecasts still cannot capture all the local scale variability of persistence of wet and dry days as well as magnitudes of rainfall on individual days and the effect of catchment antecedent moisture conditions. More local scale rainfall forecasts are thus still needed in the South African region. In this particular study the methods used did not produce convincing results in terms of runoff forecast accuracy and skill scores. The poor performance can probably be attributed to the relatively unsophisticated nature of the downscaling and interpolative techniques used to produce daily rainfall forecasts at a Quaternary Catchment scale. It is the author's opinion that in the near future, with newly focussed research efforts, and building on what has been learned in this study, more reliable agrohydrological forecasts can be used within the framework of water resources and risk management, preventing loss of life, saving considerable hardship and saving affected industry and commerce millions of rands annually. / Thesis (M.Sc.)-University of Natal, Pietermaritzburg, 2002.

För att ta fram vårflodsprognoser används idag en hydrologisk modell, HBVmodellen, som används operationellt både i Sverige och i andra länder för att ta fram tillrinningsprognoser. Ett av problemen med dessa prognoser är dock att de ofta har felaktigheter i beräkning av tillrinningsvolymer. I den här studien undersöks Vängelsjöns delavrinningsområde. Ett prognosområde där HBV-modellens tillrinningsprognos tidigare gett stora volymfel. Rapporten undersöker därför om det går att korrigera den operationella modellens snömängd genom snödjupsobservationer, för att ta fram mer tillförlitliga vårflodsprognoser. I rapporten har snödjupsobservationer och densitetsmätningar från Vattenregleringsföretagen (VRF), tillsammans med snödjupsobservationer från Sveriges meteorologiska och hydrologiska institut (SMHI), analyserats för att se om de kan användas för att förbättra tillrinningsprognosen för Vängelsjöns delavrinningsområde. Studien visar att det finns potential till att använda snödjupsobservationer för att minska osäkerheter i tillrinningsprognosen för Vängelsjöns delavrinningsområde. Det bästa resultatet gavs genom att använda tre stationer från SMHI tillsammans med två stationer från VRF. Resultatet för dessa fem stationer gav ett volymfel i uppmätt vattenekvivalent på 6 %, 7 %, 22 % och 17 % för mars månad perioden 2014-2017, vilket gav en markant förbättring mot det modellerade prognosvärdet med volymfel på 40 %, 77 %, 24 %, och 49 % för samma period. / Today, spring flood forecasts are calculated through a hydrological model, the HBVmodel. It is a model that is used operationally both in Sweden and in other countries to obtain catchment forecasts. One problem with these forecasts is that they often give errors in calculations to the inflow volume. In this study the Vängelsjö sub catchment will be examined. A forecast area where the HBV-model flow forecasting has given large volume errors. In the report it is therefore examined if the operational model can be corrected with snow measurement data. In this report snow depth observations and density measurements from Vattenregleringsföretagen (VRF) have been analyzed, together with snow depth observations from the Swedish Institute for Meteorology and Hydrology (SMHI) to see if the data can be used to improve the forecasting for Vängelsjö sub catchment. The study shows that there is a potential in using snow measurement observations to reduce insecurities in the flow forecast for Vängelsjö sub catchment. The best result was received by using three snow stations from SMHI and two from VRF. The result from those five stations gave a volume error in measured snow water equivalent (SWE) at 6 %, 7 %, 22% and 17 % for Mars during the period 2014-2017, which gave a significant improvement against the modelled forecast value with volume errors of 40 %, 77 %, 24 % and 49 % for the same period.

A previsão hidrológica possibilita a identificação antecipada de eventos de cheia potencialmente causadores de inundação, o que é de grande importância para a atuação de entidades como a Defesa Civil. Quando se deseja estender a antecedência no tempo em relação a estes eventos, principalmente nos casos onde a bacia de interesse é relativamente rápida, torna-se necessária a incorporação de previsões quantitativas de precipitação (QPF) na modelagem hidrológica, as quais podem ser obtidas a partir de modelos numéricos de previsão do tempo. Entretanto, a falta de acurácia atribuída a estas previsões de chuva, dadas de forma determinística, vem promovendo sua substituição por sistemas de previsão meteorológica por conjunto (EPS - Ensemble Prediction Systems), cuja finalidade é a geração de possíveis estados futuros da atmosfera para considerar as incertezas associadas ao seu estado inicial e às deficiências na representação física dos modelos de previsão do tempo. Neste contexto, o presente estudo teve por objetivo avaliar uma metodologia de previsão de cheias por conjunto na bacia do Taquari-Antas/RS até a cidade de Encantado (19.000 km²), localizada na região Sul do Brasil. Para tanto, foi utilizado o modelo hidrológico MGB-IPH acoplado a diferentes sistemas de previsão, sendo eles: (i) EPS Regional ETA, de curto prazo (até 72 horas) com 5 membros de diferentes parametrizações; (ii) EPS Global ECMWF de médio prazo (até 10 dias) com 50 membros de condições iniciais perturbadas, incluindo perturbação estocástica nos parâmetros de ajuste do modelo e; (iii) Previsão Determinística do Modelo Regional ETA (até 7 dias). A avaliação das previsões consistiu em dois hindcastings distintos, envolvendo uma análise visual de eventos singulares ocorridos em 06/06/2014 e 21/07/2011 além de uma análise estatística no período de Mar/2014 - Nov/2014. Durante a análise visual foi possível identificar, a partir de antecedências de 5 a 6 dias, uma persistência na previsão dada pelo crescente número de membros do conjunto de médio prazo (ECMWF) com superação dos limiares de referência, na medida em que se aproximavam os eventos de cheia. Apesar da grande incerteza na magnitude das previsões hidrológicas para o conjunto de curto prazo, a vazão máxima foi relativamente bem prevista por pelo menos 1 membro em quase todas as antecedências, enquanto que a previsão do timing dos eventos foi considerada de boa confiabilidade. Durante a avaliação estatística foi possível notar uma falta de espalhamento nos conjuntos, com tendência de subestimativa de acordo com o aumento da antecedência. Em uma comparação com previsões determinísticas, as previsões por conjunto demonstraram maior acurácia principalmente até 72 horas de antecedência, com destaque para a maior probabilidade de detecção dos limiares de referência e manutenção de falso alarme a níveis reduzidos. Além disso, verificou-se também que a agregação de previsões efetuadas em tempo anterior naquelas atuais acarreta em ligeira ampliação do espalhamento do conjunto e maiores probabilidades de detecção dos limiares de alerta para os membros mais elevados, apesar da redução no desempenho em termos de acurácia e viés. De forma geral, as previsões por conjunto apresentam potencial para servir como uma informação complementar em sistemas de alerta contra cheias, possibilitando uma melhor preparação dos agentes envolvidos durante a ocorrência destes eventos. / Hydrological forecasting plays an important role for issuing flood warnings, allowing for anticipation and better preparation of authorities at the occurrence of such events. In order to extend lead time in a flood forecast, especially when the catchment response time is relatively fast, it may be useful to couple a hydrological model to quantitative precipitation forecasts (QPF), usually obtained directly from numerical weather prediction (NWP). However, deterministic (i.e. single) QPF are usually referred to many errors and lack of accuracy, mainly caused by uncertainties on initial state of the atmosphere and on physical representation of weather forecasting models. To address these shortcomings, it becomes necessary to take into account the uncertainties associated to rain forecasts, which can be represented by Ensemble Prediction Systems (EPS). The purpose of such systems is to provide different trajectories of the atmosphere by perturbations on its initial condition and on parameterization schemes of the models, generating an ensemble of forecasts that can be used as input to hydrological modelling (HEPS). In this context, the present study aimed to assess a methodology of ensemble flood forecasting on Taquari-Antas basin up to the city of Encantado/RS (19.000 km²), located in southern Brazil. Therefore, the hydrological model MGB-IPH was coupled to different forecasting systems: (i) Short Range EPS ETA (up to 72 hours), a regional model with 5 members of different parameterization schemes; (ii) Medium Range EPS ECMWF (up to 10 days), a global model with 50 members of perturbed initial conditions and stochastic perturbation in the model parameters; (iii) Deterministic ETA Model (up to 7 days). The forecasts were evaluated by two different hindcastings, which includes a visual assessment of singular events occurred on 2011 and 2014 and a statistical analysis for the period between Mar/2014 and Nov/2014. It was possible to identify a forecast persistence on medium-range for the selected events, by the increasing number of members exceeding the reference thresholds from lead times up to 5 - 6 days. On the short range, although large uncertainties in the magnitude of hydrological forecasts were found, the peak discharge was well forecasted - at least for a single member - in nearly all lead times, whereas the prediction of the peak timing was considered reliable. Regarding to statistical evaluation, an inadequate spread in the ensemble was observed from short- to medium-range, with a tendency of underestimation for longer lead times. In a comparison with deterministic forecasts, the ensemble forecasts showed higher accuracy especially up to 72 hours in advance, including highlights on greater probability of detection (POD) above the reference thresholds even with low false alarm rates. It also was found that the a combination of previous forecasts on the recent ones leads to a slight increase of ensemble spread and POD for higher members, despite the performance reduction in terms of accuracy and bias. In summary, the hydrological ensemble forecasts demonstrated a good potential to serve as an additional information within a Flood Alert System.