Flood scenarios

The safety of nuclear power plants can be seriously affected by flooding, both for sites on rivers and for sites on the sea coast (including enclosed and semi-enclosed water bodies) or large lakes. 

Floods can be associated with either frequent or rare events, according to the definitions provided in Refs [1,4]. The procedures to be used for data collection and the methods to be used for hazard evaluation will depend to a large extent on the nature of the flood. 

The design basis flood has to be derived from the flood hazard for the site, which is a probabilistic result derived from the analysis of all the possible flooding scenarios at the site. However, in some cases the design basis flood is evaluated via deterministic methods and no probability is attached to it. In these cases a probabilistic evaluation should always be carried out to be able to compare the contributions of different design basis scenarios to the overall plant safety (see Ref. [6]) and to evaluate the overall probability of radiological consequences of a potential plant failure. 

The design basis flood is a series of parameters that maximize the challenge to plant safety as a consequence of a flood the parameters may be associated, for example, with the maximum water level, the maximum dynamic effect on the protection or the maximum rate of increase in water level 

For coastal sites (sea, lakes and semi-enclosed water bodies) the flood hazard is related to the most severe among the following types of flood, where apphcable  

---

Potential accident scenarios and flood locations were identified from plant drawings and tlic RHR system fault tree that identifies the equipment and support needed for RHR system operation. The equipment location was correlated with flood areas with consideration for plant features which may impede or divert the flow. The flood scenarios identify the effect on systems required to prevent core damage. Quantification accounts for the rate of rise of the flood relative to the critical level in each specific plant area. The time available for any recovery action is calculated from tiic volume and the flow rate. 

Ice and hail, snow, etc. can cause a loss of off-site power. The reactor building was designed to withstand blast pressure of 1,000 Ibs/ft can withstand tornado missile impact (Sharp, 1986). A tornado could damage the reactor by hitting the river water pump houses similar to the flooding scenario. 

A first-order estimate of the impacts of the flood scenario can be obtained through an evaluation of the flood depth map after it has been overlaid onto an electronic version of a standard topographic map that depicts roads, buildings, and public facilities. Such maps are useful for preparing evacuation plans and selecting locations for roadblocks. The FLDWAV output files may be consulted to determine the time characteristics of the flood, such as the time from the dam breach to the onset of flooding, and the duration of flooding. 

Excess volume or flood diversion trenching to handle a 100-year flood scenario... 

An analysis of the potential for extreme flooding scenario for the Sandia Pulsed Reactor (SPR) was accomplished in 1999 (Pickard 1999). This analysis concluded that the likelihood of rainfall sufficient to completely flood a facility, such as the HCF, was of the order of once per million years. 

To quantify flood risks, state of the art modeling techniques combine (i) probability density functions of hydraulic conditions (ii) probability density functions of the variables that determine the load bearing capacity of a flood defense, (iii) fault tree models to analyze failure modes, and (iv) flood propagation models, land-use data and loss functions to relate flood characteristics and land-use data to the consequences of flood scenarios (e.g. Van Manen Brinkhuis 2005). This paper focuses on the quantification of loss of life for a given flood scenario. The quantification of flood probabilities and flood characteristics (such as flow velocities, rise rates, and inundation depths) is outside the scope of the present paper and discussed in e.g. Steenbergen et al. (2004). 

The ninnber of fatalities for a particular flood scenario depends on the spatial distribution of (titne-variant) flood characteristics, the population densities of affected regions, possibilities for evacuation, and the probabilities of death of the affected individuals. The probability of death of an affected itidividual is typically assinned to depend only on flood characteristics, i.e. hydraulic conditions. Differences between the vulnerabilities of different individuals are thus ignored. The approach used in the Netherlands is shown schematically in Figure 3. 

The mortality functions can be used to estimate the probability of death at a given location for any given flood scenario, as well as the number of deaths for... 

The second risk metric that is being considered by policymakers is individual risk. Individual risk is defined as the probability of death of an average, improtected person that is constantly present at a certain location (note that evacuation could be included in an alternative definition). As levels of individual risk are highly dependent on local flood conditions and topography, it is troublesome to estimate individual risks throughout low-lying regions, without detailed information from flood scenario calculations available. No nationwide estimates are therefore presented for individual risk. 

Method 3 A simplified Bayesian approach Constructing a probability density function for the number of fatahties from floods would require us to compute flood probabilities and consequences for a vast number of flood scenarios. This is exactly what is being done within the FLORIS-project for all dike rings in the Netherlands. The results of that project will come available by the end of 2011. But pohcymakers needed insight into the severity of fatality risks well before that date. A simplified procedure was therefore designed to estimate societal risks from floods on the basis of only limited data. The simplified procedure rests on the following assumptions ... 

For the estimation on the mortality (Fd) flood scenarios were grouped into three broad categories. These are defined by the severity of the flood, water depths and/or flow velocities, and to what extent the flood comes unexpected, with little or no warning. The three categories are ... 

Combinations of two or more dependent events should be carefully analysed with account taken of the dependence or independence of the events. For example, on a river site exceptional spring runoff floods may cause the collapse of an ice jam resulting in higher water levels at the site and the possible obstruction of water intakes by ice floes. On a coastal site a tsunami or a storm surge may occur at the time of an exceptionally high tide. Special attention should be paid to sites on estuaries for which flood scenarios may show aspects of both coastal and river sites. 

The availability of communication routes external to the site during and after a flooding event involves facilities that are not always under the direct control of the site administrators. Since the availability of such communication routes is a key part of the emergency planning, a dedicated analysis of the flooding scenario should be performed together with the competent authorities as part of the hazard evaluation for the site. 

In some external scenarios, the development of the consequences for the facility is not proportional to the growth of the load. In these cases a cliff edge effect is recorded, i.e. a sudden increase of the consequences as a result of a small increase of the causes. A typical example is the flooding scenario in a site protected by a dam, where as soon as the water is higher than the protection, the whole site is flooded to the maximum level. 

---