Hydra-Zoet Probabilistic model for the assessment of dike heights : probabilistic model for the assessment of dike heights

C. Geerse, R. Slomp, H. de Waal ; HKV consultants, Ministerie van Infrastructuur en Milieu, Rijkswaterstaat Water, Verkeer en Leefomgeving (RWS, WVL), Deltares
[S.l.] : HKV consultants
01-12-2011

According to the Flood Defences Act [Wwk, 1996], now the Water Act 2009, every five years the Dutch primary flood defenses have to be assessed. To that purpose Hydraulic Boundary Conditions (HBC) are used (normative water levels and wave conditions). The most recent ones date from 2006 (HBC2006), and are used in the third period of assessments, corresponding to 2006-2011. A large part of the HBC are determined with so-called Hydra models. These are probabilistic models, implemented in computer programs. These models are used for the primary flood defenses of type a, which are located along the larger water systems in the Netherlands, i.e. along the rivers Rhine and Meuse and their branches and the river (Overijsselse) Vecht, along the large lakes Lake IJssel and Lake Marken and along the coast. Two important water systems are the tidal river area and the Vecht and IJssel delta. The tidal river area is that part of the lower reaches of the Rhine and Meuse where storms on the North Sea have a significant effect on the water levels during high discharge waves. The Vecht and IJssel delta consists of the lower reaches of the Vecht and IJssel (a branch of the river Rhine) where wind set up from Lake IJssel has a significant effect on the water levels during high discharge waves. For the tidal river area the model Hydra-B is available, and for the Vecht and IJssel delta the model Hydra-VIJ. These models have been developed by Rijkswaterstaat, a department of the Ministry of Transport, Public Works and Water Management, and HKV Consultants from 1999 until the present. There are a lot of reports regarding these models, but due to the long period over which the models have been developed, the reports are sometimes inarticulate and incomplete. It is therefore desirable to provide a coherent description of these models, including statistical data and modelling, physical modelling and probabilistic formulas. Providing this description is one of the main purposes of this report. The models Hydra-VIJ and Hydra-B have many similarities. However, in the past years it became apparent that these models can be constructed in a more uniform way. All the so-called “slow” stochastic variables such as discharges and lake levels are modelled using the same type of schematisation. This resulted in a single probabilistic model for the lower reaches of the Rhine, Meuse, Vecht and IJssel. Even better, the upper reaches of these rivers and Lake IJssel and Lake Marken fit into the general scheme of this model as well, meaning that all flood defences of type a of the fresh water systems are part of a single new probabilistic model, called Hydra-Zoet. The model offers big advantages in terms of clarity, management and maintenance. The second main purpose of this report is to describe the detailed formulas of Hydra-Zoet. The models Hydra-VIJ and Hydra-B contain several calculational options and failure mechanisms. For transparency, this report only treats the major options and failure mechanisms of these models and of the model Hydra- Zoet: overflow and wave overtopping, used respectively in the calculation of normative water levels and required crest heights, corresponding to exceedance frequencies given by law. These concepts play an important role in the 5-yearly assessment of the flood defences. Other options and failure mechanisms, such as the determination of wave conditions to assess dike revetments, are left out of the description. The report contains detailed statistical and mathematical formulas for the models Hydra-VIJ, Hydra-B and Hydra-Zoet. It also discusses (but in less detail) the physical models for waves and water levels and the way they are used to generate input for the Hydra models. Lastly, the report briefly describes the role of the Hydra models in “the chain of assessment, design and policy” regarding flood defences. The mathematical parts of the report require a fair amount of statistical and mathematical background, whereas the report as a whole can serve as a reference book.

159 p.
In opdr. van RWS, WVL
With ref.
Volgens de Wet op de Waterkering [Wwk, 1996] nu Waterwet 2009 moeten de primaire waterkeringen iedere vijf jaar worden getoetst. Daarbij worden Hydraulische Randvoorwaarden (HR) gebruikt. De meest recente HR zijn die uit 2006 (HR2006), die gebruikt worden in de derde toetsronde, die de periode 2006-2011 beslaat. Een groot deel van de HR-onderdelen wordt bepaald met zogenaamde Hydra-modellen. Dat zijn probabilistische modellen, die zijn geïmplementeerd in computerprogramma’s. Deze modellen hebben betrekking op de primaire a-keringen. Voor de benedenrivieren is er het model Hydra-B en voor de Vecht- en IJsseldelta het model Hydra-VIJ. De modellen Hydra-B en Hydra-VIJ zijn ontwikkeld door Rijkswaterstaat, in samenwerking met het bureau HKV Lijn in Water, gedurende de jaren 1999 tot heden. De rapportage over Hydra-B en Hydra-VIJ is zeer uitgebreid, maar gezien de lange ontstaansgeschiedenis en het grote aantal rapporten, soms wat samenhangend en een enkele keer incompleet. Het is daarom gewenst een samenhangende beschrijving te geven van beide modellen, waarbij statistische gegevens, fysische modellen en probabilistische formules aan de orde komen. Het geven van deze beschrijving is één van de hoofddoelen van dit rapport. De modellen Hydra-VIJ en Hydra-B vertonen de nodige verwantschap. In de afgelopen jaren is echter duidelijk geworden dat de modellen op nog meer uniforme wijze kunnen worden opgezet, waarbij de modellering van de zogeheten “trage stochasten”, zoals afvoeren en meerpeilen, op overeenkomstige wijze kan worden uitgevoerd. Na deze uniformering zijn de benedenrivieren en de Vecht- en IJsseldelta deel gaan uitmaken van een nieuw probabilistisch model Hydra-Zoet.
Bovendien is gebleken dat ook het IJssel- en Markermeer en de bovenrivieren kunnen worden opgenomen in dit nieuwe model. Dat betekent dat alle primaire a-keringen uit de zoete watersystemen dan deel uitmaken van één nieuw probabilistisch model, Hydra-Zoet genaamd. Dat laatste biedt grote voordelen qua overzichtelijkheid en voor beheer en onderhoud. Het tweede hoofddoel van dit rapport is het beschrijven van de formules uit Hydra-Zoet.