Engineers are nowadays expected to not only be skilled in technical matters, preparing calculations and designs, but also to know far more about risk analysis than ever before. In almost all aspects of engineering, risk analyses are now commonplace, with many of the design codes used based on statistical analyses. In the fields of river and coastal engineering it is nowadays expected that risk analyses will be undertaken as a standard element of a project. Part of this analysis is to determine the frequency of when particular events might occur. But the most complex and challenging part is to determine what the consequences of such events would be. Engineers make use of complex modeling programs to simulate these consequences, and one of the most versatile of these for simulating flooding is the InfoWorks RS program.
Many low lying areas of land alongside rivers or along the coast have flood defense embankments, walls or levees that prevent these areas being flooded for up to specific river or tide levels. There are rising sea levels in many areas around the world, and in time many existing flood defenses will be overtopped during storm or surge tide conditions. Deciding when these flood defenses need to be raised and by how much is a challenge which faces many of the flood defense authorities. Many of the defenses are old and another part of the challenge is to elongate the life of these structures. A key element of this is to determine the consequences of the existing defenses failing or a breach occurring. InfoWorks RS can simulate any form of breach, and this enables the Engineer to simulate the flooding which would result. An essential part of this analysis is to model the exact configuration of the breach that might occur, and that is the focus of this article.
Digital Terrain Models
In order for the model to adequately simulate the flooding across flood plains and across any defended areas (should a breach occur) it is essential that a digital terrain model underlies the full model. Digital Terrain Models can be created from data acquired in a number of different ways and to differing standards, and can either be gridded data or triangulated data. There is some debate about whether it is best to use Digital Terrain data (also known as ‘bare earth’ data) without any buildings or to use Digital Elevation data, which includes buildings and tree canopies.
There are a number of airborne techniques for acquiring digital terrain and digital elevation data, one of which, Lidar (Light detection and ranging), can also be used from helicopters. Helicopters can fly at low altitude along just the line of the flood defense structures, allowing very high accuracy data acquisition. There are a number of authorities that now use this technique on a regular basis to check the integrity of earth embankment defenses, because any slumps, slips or failures can be quickly identified. Obviously, should any potential embankment failures be identified it will then be necessary to establish what the consequences would be of a breach at that location. A key element in this analysis is to model the shape, or configuration, of the breach accurately.
It is possible to model almost any breach configuration, but examining too wide array of possibilities makes it difficult to progress. What is needed is a logical approach to the configuration of the breach. The key factor is whether or not the existing defenses are ‘hard’ (eg walls) or ‘soft’ (eg earth embankments) because in general ‘hard’ defenses once breached are less likely to be subjected to scour, whilst ‘soft’ defenses will continue to scour after the initial breach. To take this further, it is also necessary to establish whether overtopping of the defenses is likely to occur. Overtopping of ‘hard’ defenses may not lead to a breach, but overtopping of ‘soft’ defenses may cause scour of the embankment and this scour would eat away at the embankment until a breach occurs which, then becomes even more severe as more water causes more scour.
Propagation of the breach along the line of the defenses also becomes important. With ‘soft’ defenses it is frequently the case that the breach will widen as the ends of the remaining embankments are scoured away, with the speed of propagation depending on the flow velocities and the amount of material to be scoured away. With ‘hard’ defenses the initial breach may have been due to failure of a particular section of wall and it is not necessarily the case that adjacent sections of wall will also fail. It maybe that the failed section was built differently or may have been defective in some way.
InfoWorks RS allows the breach to be defined easily in terms of the geometry at different times during the simulations. The geometry is very simple to define, with 5 variables including breach location, width of breach, the invert or base level of the breach, the geometry of the base of the breach and the side slopes at the ends of the breach. In most cases breaches in ‘hard’ or ‘soft’ defenses would be defined differently - with ‘hard’ defenses the breach geometry would not change greatly during the flood event, whereas with ‘soft’ defenses it is likely that the breach would widen and deepen progressively during the simulation.
Simulating Repairs to Breaches
There are many instances where the consequences of flooding due to a breach can be significantly reduced if the breach is repaired within a reasonably short timescale. In areas with very long flood defenses it may be more cost effective to have resources and materials available at local depots to repair breaches quickly rather than spending capital investment to replace or upgrade ageing defenses. This is particularly the case with sea defenses, when the flood risk is solely at high tide, allowing access to the breach at low tides, albeit for short times, to effect repairs.
In the same way that the formation and propagation of breaches can be simulated it is also possible to simulate how breaches are repaired. This is a simple matter of continuing to define the breach geometry for the different time intervals, but this time the breach width is reducing and/or the base level is rising. If the times are varied in different versions of the model it is possible to evaluate the extent and cost of flood damage with different breach closure times. Having established what repair period is required on the criterion of cost, the next step could be to evaluate how easily that could be accomplished and what resources would be required.
InfoWorks RS provides a very flexible tool for the Flood Defense Engineer to undertake detailed and comprehensive risk analyses in relation to the integrity and adequacy of flood defenses. The adequacy of the defenses can be assessed on the basis of whether defenses at different levels are overtopped or not. As far as the integrity of the defenses is concerned, InfoWorks RS allows this to be done in a systematic manner. Breaches at different locations can be modeled and these could be simulated as being triggered by, for example, the sudden collapse of a wall or the erosion of an earth embankment due to overtopping. Widening and deepening of the breach during the flood event can be modeled, and the consequences of flooding in the defended areas can be fully simulated.
InfoWorks RS also allows the engineer to consider alternatives to expensive capital investment to replace ageing defenses. The effectiveness of having resources and locally stockpiled materials to repair breach within set time periods can be evaluated, perhaps modeled as a series of “what if” scenarios, with the associated flood damage costs for each scenario quickly established from the modeling.
Whilst the simulation of breaches in sea defenses and fluvial defenses in relatively wide rivers can be undertaken without difficulty there are a small number of special cases where something else might also be required. These special cases are typically in elevated canals and narrow elevated watercourses (eg mill races) where the width of the canal or watercourse might be narrower than the width of the breach; in these cases it might be the geometry and capacity of the canal or watercourse which governs the flows discharged rather than the geometry of the breach. Modeling of these special cases is more difficult, especially if there is a significant degree of scour and erosion associated with the breach. These special cases tend to be unique and there are a number of ways in which they can be satisfactorily modeled. The simplest way, in the case of canals, is to use the multitude of bridge locations (where the canal narrows under a bridge) as a means of controlling or limiting flows. It also tends to be these same locations where stop logs are inserted to stop the flow of water. These can be modeled in InfoWorks as sluices with a Real Time Control which replicates the stop logs being inserted.