Flood prevention schemes: guidance for local authorities

SECTION 4: ASSESSING THE BENEFITS

4.1 EVALUATION OF LOSSES

Domestic and other urban property

4.1.1 Permanent buildings at risk of total loss from flooding should usually be valued at their current market value, excluding any adjustment in value for the flood risk. For many strategic and preliminary studies, the mid-range of council tax bands, suitably adjusted from their 1991 price datum, can be used to estimate property values. Generally, property will be assumed to be written off if the mean annual maximum water level exceeds the floor level.

4.1.2 However, market values sometimes need adjustment. For example, for properties such as pubs and restaurants, the market value includes a significant factor for customer goodwill. This 'goodwill' element is a transfer, not an economic loss (Annex A). In other circumstances, if there is an excess supply of, say, some types of commercial property, such property would not be replaced if lost, and no economic loss would be incurred. It is also important to avoid double counting any waterside amenity element of market value. Further, in the case of loss through abandonment, it should be assumed that the contents of the buildings are removed before the building is lost. Consequently, all removable fixtures and fittings should be excluded from the valuation.

4.1.3 Where a flood has occurred recently, a record of the damages incurred will often be a good starting point for evaluation of losses. However, it is important that the limitations of such data are clearly understood. Actual damages will rarely be available for the required range of events. Where losses are recorded, these will often be in financial rather than economic terms, and values will have to be converted.

4.1.4 The Flood Hazard Research Centre produces some standard data on the losses to be expected for different types of property, according to the type of dwelling, its age and other aspects. Adjustment factors, which take account of the additional losses from saline flooding, are also available. Details on these data and factors can be found in the manuals listed at references 3-5. The manuals have been combined in reference 6, which is currently being reviewed.

4.1.5 For large industrial or commercial properties, or unusual properties such as listed buildings, it will often be necessary to carry out a site survey of the likely losses. A questionnaire can be used for this purpose (reference 5). In times of rapid obsolescence and replacement of many commercial and industrial buildings, effective life should be carefully considered when assessing the value of such properties.

4.1.6 An alternative to standard data is to commission surveys by loss adjusters. However, this is expensive. It is rarely justified unless the properties concerned are atypical and the use of standard data would likely give misleading results.

4.1.7 From the foregoing:

• land and buildings should be valued in constant real prices in their current use;
• any 'goodwill' element in values for commercial premises should be excluded; and
• care should be exercised in the derivation of non-standard valuations.

Temporary and semi-permanent structures

4.1.8 There are other cases where the real economic value of losses may be very different from current market values. For example, the economic value of a mobile home on a particular site is equivalent to the cost of moving it there and establishing the site, not the value of the unit itself, which could be retained if it were relocated elsewhere. Also, in assessing economic damages, caravans, mobile homes, chalets or other temporary buildings or structures should be considered as depreciating assets worth, on average, only half their replacement cost.

4.1.9 For the 'do nothing' case, it should normally be assumed that any caravan or mobile home could be relocated. The economic loss would then be limited to the cost of removal together with the loss of installed infrastructure, depreciated as appropriate. Where a site is to be protected, the 'do something' damages should be calculated in the normal way, taking into account the seasonal nature of occupation. Similar considerations will apply to other temporary or relatively short-life structures, such as most amusement park rides.

Infrastructure

4.1.10 The market value of a property can be expected to include the value of the immediate services; such local services have no economic value once the property is lost. However, separate valuations will usually be appropriate for infrastructure serving a wider area, including trunk sewers, main road and rail routes, major pipelines, cables, pumping plant and other such facilities. In general, the loss of such infrastructure can be treated as the cost of replacing the facilities elsewhere or rerouting them. Appropriate adjustments should be made for depreciation and obsolescence.

4.1.11 Embankments constructed primarily as flood defences have a functional value only in terms of the protection that they provide. Including a value for such assets is likely to lead to double counting. However, they may also have a use value for recreation, which may be taken into account.

Indirect losses

Consumer losses

4.1.12 If a shop or factory is flooded, the company will lose sales and its customers may be inconvenienced. Potentially, therefore, there are 2 forms of indirect losses: to the consumer, and to the supplier. In general, the loss to the consumer is the economic loss; that to the supplier is usually financial rather than economic. If consumers can buy the same goods at the same cost from an alternative supplier immediately, there is no loss to them. If they have to make do with inferior goods or incur higher costs, there may then be an economic loss. However, it will only be appropriate to evaluate this in special circumstances; for example, where long-term loss of a rural retail outlet is likely to involve significant extra travel.

Supplier losses

4.1.13 On the suppliers' side, if other shops or factories make up the consumers' purchases, this is simply a transfer unless those other shops or factories incur higher costs (Annex A). The sales lost by one company are gained by another. The only exception is when those purchases are made up by additional imports or lost exports. Indirect losses do not normally arise from disruption to commercial and retail activity because there are typically many alternative outlets offering the same services immediately. This need only be considered in exceptional circumstances, for example when highly specialised products are involved.

Traffic disruption

4.1.14 Disruption of road and rail networks, can result in significant indirect losses. For road networks, it will not generally be worth evaluating these unless a major through road is closed by at least the 1 in 10 year flood. If flooding occurs below the 5-year event, and a significant part of the network carrying through traffic is affected, the benefits of reducing disruption can be large, both in total and as a proportion of scheme benefits. Traffic that usually uses the roads will have to divert, and may have to travel further, and/or for longer, incurring both resource and time costs. Since the speed of traffic depends on volume, the normal traffic on the diversion routes will also travel more slowly, again increasing such costs.

4.1.15 Methods and guidance are available to help calculate the difference in the resource and time costs of using the road network under different flooding conditions, and to address the special problems of calculating the costs of flood-induced traffic disruption.

4.1.16 One problem is that of identifying the diversion routes. Another is that the progressive development of flooding may induce a cascade of traffic diversions as one road after another is closed. Further, standard volume-speed relationships are not intended for highly congested traffic, and their blind application can yield results which are not strictly applicable.

4.1.17 Where traffic disruption is likely to be severe and extend over a prolonged period, it may be more realistic (and more cost-effective for the appraisal) to equate the economic loss to the cost of reconstructing the road, or making sufficient improvements to alternative routes to avoid the cost of delays. This should be applied only where the present value is likely to be less than that of the long-term costs of disruption.

Socio-economic equity

4.1.18 A flood prevention scheme might have differential impacts on individuals, depending on aspects such as their income. It may therefore be necessary to consider the question of social equity. This can be achieved through a 'Distributional Impacts ( DI)' analysis, which examines the distribution of costs or benefits of interventions across different income groups and social classes (reference 2). Interim advice on this topic has been produced by Defra (reference 7), and is summarised in paragraphs 4.1.19-4.1.23.

4.1.19 If a decision is made to assess DI, appraisers should be aware of the principle of diminishing marginal utility of additional consumption, whereby the impact of a project on an individual's wellbeing may vary according to his or her income; the rationale being that an extra pound will give more benefit to a person on a low income compared to someone on a high income. In other words, as income rises, its marginal value reduces. Consequently, a loss of £1000 will matter more to someone on a low income. For flood prevention schemes, DI analysis can be applied to the evaluated costs of avoided damage to residential property. The subsequent costs arising from the analysis may then be treated in the conventional manner.

4.1.20 The Treasury Green Book recommends that DI should be applied where it is necessary and practical to do so. Determining if it is 'necessary and practical', depends on a number of circumstances, including (i) whether a community at flood risk can be identified with reliable data and categorised according to their prosperity or social class; (ii) whether the assessment will contribute to an appraisal that demonstrates equity and fairness to people; and (iii) whether the time and effort in undertaking the assessment is proportional to the scale of the overall appraisal, either at a strategic or scheme level.

4.1.21 In addition, appraisers should consider whether they feel that in not undertaking the assessment, a strategy or scheme will still have an adverse differential impact on a particular group. A decision not to adjust explicitly for distributional impacts will require to be justified.

4.1.22 The following 2 steps set out the procedure for transposing the guidance on DI into flood defence investment:

Step 1 Analyse and understand the level of knowledge on the type, age and number of residential properties; the mix of social class groups and levels of income within an appraisal area. Take account of DI by following Step 2, if necessary and practical. If it is not necessary and practical, ignore Step 2 and use standard depth damage curves that focus on property type and age, only, without accounting for social class mix or income level.

Step 2 If proven necessary and practical, and good quality information is obtainable, Total Weighted Factors may be used by social class group as shown in Table 4.1. Those for social class groups C1 and C2 will generally have a negligible effect on the DI assessment. Hence, this approach is only recommended where AB or DE social class groups are predominant. The factors may then be applied to adjust the standard depth-damage curves, to obtain damages avoided taking account of DI.

Table 4.1: Total Weighted factors

Total weighted factors by social class group
AB	C1	C2	DE
0.74	1.12	1.22	1.64

In the interest of transparency, both weighted and unweighted results should be routinely presented. Where results are sensitive to any weighting adjustment, a sensitivity analysis should be provided.

4.1.23 Where the quality of available information permits, appraisers should take account of DI in homogeneous areas or areas with a high proportion of rented accommodation. For the latter, the income level or class of the owner of the property should be used for assessing building damages, and that of the occupier used for the contents damages. The approach in Step 2 is suggested, but again only if a DI assessment is necessary and practical.

Non-monetary impacts on households

Stress

4.1.24 To householders, impacts of flooding such as increased stress, health effects and loss of memorabilia, can be as important as the direct material damages to their homes and their contents. Research has provided a method for quantifying these human related intangible impacts, as indicated in the recent interim guidance note produced by Defra (reference 7). In summary, the note states that the value of avoiding the health impacts of fluvial flooding is of the order of £200 per year per household. This is a weighted average derived from a wide range of responses. The note also provides a risk reduction matrix which can be used to calculate the value of health related benefits for different standards of scheme protection. The matrix is reproduced in Table 4.2. For example, the highlighted figure of £188 in the table represents the annual health related damages avoided, and hence the benefit per annum, per household, in moving from a pre-scheme situation where the standard of protection is 1 in 20 years to a scheme with a 1 in 100 year design standard.

Table 4.2 Intangible benefits associated with flood defence improvements (£ per annum per household)

			Standard of protection after scheme - in terms of annual flood probability and (return period in years)
Standard of protection before scheme - in terms of annual flood probability and (return period in years)			0.007 (150)	0.008 (125)	0.010 (100)	0.013 (75)	0.020 (50)	0.033 (30)	0.05 (20)	0.1 (10)
	1	(1)	£218	£215	£200	£153	£73	£25	£12	£5
	0.1	(10)	£214	£210	£195	£148	£68	£21	£8	£0
	0.05	(20)	£206	£202	£188	£141	£60	£13	£0
	0.033	(30)	£193	£189	£175	£128	£47	£0
	0.020	(50)	£145	£142	£127	£80	£0
	0.013	(75)	£65	£62	£47	£0
	0.010	(100)	£18	£15	£0
	0.008	(125)	£4	£0

4.1.25 For areas of uniform flood risk, such as housing on level ground, damages are based on common standards of defence. Having identified the standards of protection before and after an option is implemented, Table 4.2 can be used to derive the annual intangible benefit per household. This can be applied to the total number of households (or residential properties) in the area to give the overall annual intangible benefits for a particular option. In areas where the risk varies greatly, such as sloping ground away from a river, damages are based on individual levels of property flood risk. This will require banding of the areas into different levels of existing protection, and the identification of the standards of protection being offered by the scheme to each band area. The table can then be used to evaluate the intangible benefits for the properties within each band.

4.1.26 The figures in Table 4.2 already take account of distributional impacts. Consequently, a DI analysis should not be applied to the results arising from the appraisal of intangible impacts.

Disruption

4.1.27 A partial measure of the disruption resulting from flooding can be given by the cost of renting a home equivalent to the one flooded, together with the cost of accelerating the drying-out process.

4.1.28 About 50% of households are vacated for an average of 30 days when flooding exceeds 30 centimetres (reference 1). The use of dehumidifier units is the best method of drying out properties, and hence of enabling repairs and redecoration. The rental of such units reduces the losses that would otherwise occur. The number of dehumidifiers required depends on the size of the property. Rental and electricity rates are used to calculate the costs. On average, 2 to 3 units are required for a period of 3 to 4 weeks.

4.1.29 If there are reasons to believe that the remaining non-monetary impacts will be unusually high, this aspect should be considered. For example, if a flood were to last several days or result in a significant risk to life, stress and health damages may be anticipated to be higher than average.

4.2 RECREATIONAL AND ENVIRONMENTAL VALUES

Recreation

4.2.1 Flood prevention schemes may affect the value of a river or coastal reach for recreational uses, including angling and informal recreation. Any significant associated gains or losses should, as far as possible, be included in the benefit-cost analysis. Where only marginal changes in recreation or amenity are likely, such valuations will seldom be worthwhile. If a scheme relies on a substantial element of recreational benefit for its justification, it should be treated as a multi-functional project (paragraphs 2.13-2.15).

4.2.2 The benefits of avoiding a loss in recreational value, or an increase in such value, can be calculated by (i) estimating the number of visits made to the site, and (ii) multiplying this by the change in the value of enjoyment per visit.

4.2.3 Thus, the number of visits made to a site is a primary indicator of the likely magnitude of the benefits. If there are no visitors, there are no benefits. Therefore, any analysis should start with a preliminary estimate of the number of visits made by adults. For rivers, where the number of visits is small, the estimates given in, for example, the Foundation for Water Research Manual on the benefits of surface water quality improvement can be used to estimate the order of magnitude of the recreational benefits (reference 8).

4.2.4 Where no comparable estimates of the annual number of visits to different types of site have been compiled, the simpler methods shown in Table 4.3 will have to be used. Early thought should be given to accurate estimation of the number of visits. The methods shown in the table are in decreasing order of accuracy.

Table 4.3 Methods of visitor estimation

Method	Comments
Long period counts using people counters	A number of infrared, or other automatic, counters are installed at least over the period of March to September in one year. The counters are manually calibrated; interview surveys are conducted to determine statistically how the number of adult visits relates to the number of passages recorded on a given day. An annual growth curve is then used to derive an estimate of the total number of adult visits made in that year.
Short period count	Counts are undertaken by hand over a period of days. An annual growth curve is then used to derive an estimate of the total number of adult visits made in that year.
Inferred estimate	The counted number of visits made to a related site (eg a car park or museum) is used to infer how many visits are made to the site. This requires estimating what proportion of all visitors to the site also visit the site for which counts are available.
Visitor equation	A number of equations have been developed which predict distance-frequency functions so that from census data on the population in different zones, a prediction can be made as to the number of visits generated by a site.
Informed estimate	The estimate of an informed person (eg car park attendant, park ranger) as to the number of adults visiting the site.
Average number of visits to equivalent sites	This benefit transfer approach is only suitable for prefeasibility and strategic studies. The number of adult visits made to the site is estimated as being of the same order as the number of visits made to an equivalent site. However, there are few sites for which good data are available and little research to enable the reliable identification of an 'equivalent' site.

4.2.5 Where recreational value is a significant part of the total benefits, a contingent valuation study may be necessary to derive a site-specific value of enjoyment. Some of the considerations for such studies are listed in Annex B. However, the difficulties and expense should not be underestimated.

Environmental and heritage issues

4.2.6 Most environmental assets have both use value (eg as measured through recreational use) and non-use values (eg existence value). These are described in more detail in Annex B.

4.2.7 Whether valued in monetary terms or not, environmental impacts should always be assessed. While in many cases sensitively designed schemes can make a significant contribution to the environment, there will often be choices to be made. It is probable that all the 'do something' options and the 'do nothing' option will have significant environmental consequences, positive or negative. Consequently, it is important to hold early consultations to establish environmental requirements and, in particular, the legislative 'drivers' (see Chapters 2 and 7). In some cases, protecting one site may have consequences for another, and a decision will have to be made on their relative values. In other cases, environmental losses may be unavoidable, for example, in reducing the risk of loss of life. However, an auditable record of the assessment and decision-making process will be required.

4.2.8 Where such choices have to be made, and in a number of other situations, it is desirable to place an explicit economic value on an environmental site or asset, in addition to any associated recreational value.

4.2.9 However, it should be recognised that it may not always be possible to express use and non-use values in monetary terms, or even to quantify them meaningfully in some other way. This should be taken into account in the overall appraisal.

4.2.10 In general, the least contentious and lowest cost method of deriving a proxy for the lower bound economic value of an environmental or heritage asset, gained or lost as a result of a flood prevention scheme, can be taken as the lowest of:

• the cost of creating a similar site elsewhere of equivalent environmental value;
• the cost of relocating to another site (eg historic buildings or relocation of specially protected species);
• the cost of local protection (eg a local flood embankment).

Creation of equivalent sites

4.2.11 The use of such proxy values, however, will depend on there being broad agreement that the value of the asset in question is at least equal to the lowest of these figures. The cost of re-creation, or removal, should be the expected present value economic cost only of those actions required to acquire the site and make the necessary changes. The normal expectation should be that long-term management costs would not change. However, such costs may need to take account of the likely success in achieving replacements of comparable quality and the time delay in their achievement. This is only strictly possible where assets are technically replaceable although, where protection can be amply justified on this basis alone, there is likely to be little benefit in applying more sophisticated valuation techniques.

4.2.12 For sites whose conservation is enjoyed by local populations only, the most appropriate valuation approach for benefit-cost analysis may be to use the value of the nearest equivalent local land in commercial use which could, in principle, be adapted as a replacement site. For example, grazing pasture or commercial woodland might provide proxy values for a local nature reserve. Clearly, this may not capture the whole value to the local community, and will not necessarily be appropriate for other decision-making purposes.

4.2.13 In certain exceptional circumstances, such economic valuations may not be appropriate. For example, an unusually high value may be placed on a feature specifically because of its particular position; or there may be some doubt as to whether the heritage value of a structure could really justify the high cost of its relocation. In such cases it may be necessary to obtain a valuation using other monetary based techniques as described in Annex B, or to apply other decision-making techniques.

4.2.14 Nationally and internationally important sites and habitats, such as SSSIs, SPAs, SACs and Ramsar sites, may be considered to have a national economic value. The lower bound estimate of this should be calculated from the minimum cost of protecting the site in situ or, if lower, the cost of providing replacement habitat. It is important to stress that calculating the replacement cost does not imply that habitat replacement is the most appropriate option. It is simply a way of deriving what can be considered to be a minimum monetary estimate of the loss involved, or the benefit of protected habitat. In some cases, habitat replacement may be necessary, but for European sites in particular, there should normally be a presumption in favour of in situ protection of habitats.

4.2.15 For some sites it will be possible to put a non-use value on the benefits of protection. Methods of calculation are described in Annex B. However, this is likely to be difficult, expensive and potentially contentious, and will rarely be justified.

4.2.16 Where there are apparent environmental disbenefits to specific options, the least contentious method of valuation may be to include, as a cost, the present value of introducing compensating environmental enhancements, which most closely match those lost.

4.2.17 Figure 4.1 shows a decision tree, using the above approach, for the determination of minimum economic values in relation to environmental and heritage sites. Other decisions will relate more to marginal changes in environmental quality or individual environmental attributes, and these may require some adaptation of the approach. However, the general principles should be applied where possible. These issues will be discussed further in Chapter 7.

Figure 4.1 Decision tree for appraising proxy economic values of environmental and heritage sites at risk of loss or damage

4.3 FLOOD PREVENTION BENEFITS

4.3.1 The benefits of flood prevention schemes are calculated as the difference between the expected value of flood damage with the option being evaluated, compared with flood damage and losses in the 'do nothing' case. A summary of the procedural steps required is shown in Figure 4.2.

Figure 4.2 Framework flow chart for flood prevention appraisal

Damage calculation

4.3.2 The damages caused by a flood are a function of its depth, duration, velocity, and its sediment and pollutant loads. In the UK, floods are usually relatively short in duration and involve low velocities. Consequently, the primary determinant of the losses for a particular property is the depth of flooding. However, in some small flashy catchments, and where protective structures fail, flood velocities can be high. This may result in additional losses from partial or complete structural failure of properties.

4.3.3 Sediment, debris and sewage borne by a flood may affect the costs of cleaning up after the event. Further, flooding by saltwater generally causes more damage than the equivalent depth of flooding by freshwater.

Calculating the benefits

Loss-probability relationship

4.3.4 The benefits of flood prevention are calculated as the expected value of annual flood losses averted. As floods are assumed to be random events, it is not possible to predict when they will occur. The expected value of annual flood losses is calculated as the probability of a range of events multiplied by the loss that such an event would incur. In practice, the losses are measured by the difference in the areas under loss-probability curves for the 'do nothing' and with scheme options. This difference in area is the expected value of the reduction in flood losses each year over the life of the scheme; the average annual benefits (Figure 4.3). These are discounted over the life of the scheme to give the present value of the benefits.

Figure 4.3 Determination of average annual benefits

Changes over time

4.3.5 Studies suggest that our climate is changing, and that this could have a negative impact on future coastal and river flooding, through rising sea levels and changing rainfall patterns. This is discussed more fully in Chapter 4. As indicated there, the projected impacts have high levels of uncertainty and, until more definite information is available, they are best examined as part of the overall sensitivity analysis (paragraphs 6.10-6.14). Further guidance on dealing with climate change risk is given in Chapter 6.

4.3.6 Where changes are anticipated in the expected probabilities of flooding over the life of a scheme, it is necessary to calculate a number of different average annual benefits corresponding to the different conditions. As noted above, such changes could include those associated with climate. Physical changes in the catchment could also affect probabilities. Further, changes over time in the use of the flood plain may alter the losses expected from a flood of a given magnitude. In such cases, average annual benefits should be calculated for appropriate years, and values interpolated for intervening periods.

The upper limit to losses

4.3.7 Care should be exercised where the total present value of losses exceeds the current write-off value of the asset. In the case of domestic or commercial property, it will usually be prudent to assume that the long-term economic loss cannot exceed the current capital value of the property. In the case of other assets, such as roads, railway lines, pipelines or cables, some very large values can be generated for long-term disruption. It will often then be reasonable to assume that the maximum economic benefit derived from flood defence is equal to the economic cost, depreciated to allow for the age of the existing asset, of reconstructing an equivalent facility at a higher level, or on an alternative alignment, which avoids the flood risk.

Sampling the return periods

4.3.8 The loss-probability curve is generally calculated using only a very small sample of the possible return period events that might be considered. The overall form of the curve, and the area under it, is derived by drawing straight lines between the calculated points. This can, potentially, result in wrong estimates of the area under the curve (Figure 4.4). In this illustration, the choice of return periods, when compared to the 'true' relationship, has resulted in a significant overestimate of the overall losses.

Approximation of loss curves

4.3.9 Determining how many and which return periods to include is a sampling problem. The aim is to obtain a reasonably close approximation to the loss-probability curve representing an infinite number of return period events if these were to be modelled. The ideal return period events to use are those located at discontinuities on the curve.

High probability events

4.3.10 First, it is important to locate the return period of the threshold flood event: that is, the most extreme flood that does not cause any damage. Secondly, engineering judgment should be used to assess where the discontinuities are likely to be. For example, they can be expected to occur when an existing natural or man-made structure is overtopped, or a culvert or bridge reaches its capacity. Thirdly, the greatest proportion of benefits generally arises from the shorter return period floods. Consequently, the sampling should usually be biased towards these events. A few judiciously chosen events at appropriate points of discontinuity will generally produce a more realistic result than a larger number at standard intervals.

4.3.11 If a software package is being used to calculate the flood losses, it is simple to plot a flood-stage/damages curve for a large number of flood stages. Discontinuities on the curve indicate those events that should be included, provided the information that produced the discontinuities is accurately provided in the data input.

4.3.12 In terms of good practice:

• the benefits should be calculated using a minimum of 3 events and the choice of those events should be considered carefully;
• one of these events should normally be the threshold flood event.

Figure 4.4 Accuracy of estimation of the loss-probability curve

Above-design-standard benefits

4.3.13 The notional standard of protection (design standard) afforded by a flood prevention scheme will usually be defined in terms of the onset of significant losses. However, many such schemes will have some effect on the losses from all floods, even the most extreme, and all of these impacts should be taken into account. While it may not always be practical to model the extent of flooding from all events up to the probable maximum flood, it should be possible to draw logical inferences as to how the scheme will respond to such larger events. From this, the likely shape of the loss-probability curve can be estimated. It is important to ensure that the range of events considered is appropriate for fair comparison of all options.

4.3.14 For example, schemes that increase the capacity of a river channel will result in less water flowing out of the bank for all events with the scheme than without. Consequently, the losses from any particular event with the scheme, should never exceed those without the scheme, and will normally be less. Two examples are illustrated in Figures 4.5(a) and 4.5(b). In both cases the shaded areas represent the total average annual benefits.

4.3.15 For other schemes, for example those involving walls and embankments that may be overtopped, losses in less probable events can be more severe than if no scheme existed. The duration of flooding may be increased, or the velocities of flow resulting from a failure may be greater than from the natural rate of rise of the flood (Figure 4.5(c)). In this case the negative benefits above the design standard should be subtracted to derive the net average annual benefits.

Figure 4.5 Estimation of above design standard benefits

4.3.16 For some schemes, above-design-standard benefits can be a significant proportion of total benefits. They will also have an impact on the incremental benefits of different design standard options. For instance, part of the incremental benefits of a nominal 100-year return period scheme may already be realised in the benefits of the 50-year return period scheme. This illustrates the importance of considering the full range of benefits in all decision-making.

Development benefits

4.3.17 Any benefits arising from potential new development, including the intensification of existing uses, should normally be excluded unless construction has commenced. The primary reason for this is to ensure that schemes are designed primarily to protect existing buildings and their users from flood risk rather than large areas of undeveloped brownfield land. Where works are proposed for economic regeneration or similar purposes, other sources of funding may be available.

Freeboard

4.3.18 Freeboard should only be used to take account of uncertainty in scheme performance. The use of indiscriminate standard freeboard allowances can lead to problems, particularly in relation to differential standards of protection. Ideally, defence crest levels should be determined from a risk-based approach (Chapter 6). It will then be possible to extend the risk analysis into the economic appraisal to determine the benefit of raising or lowering the actual height of defences.

4.3.19 In the absence of a rigorous risk analysis, appropriate methods should be used to determine a most likely range of defence levels for each return period. Where the height of the defence (including freeboard) is then designed to accommodate such uncertainty, the benefits of the defence should be those appropriate to the calculated design standard.

4.3.20 For example, if a defence is constructed on the basis of a 100-year standard, but with an additional crest height to allow for uncertainty in the hydrology and hydraulic analysis, the only benefits to accrue should be those appropriate to the 100-year standard (ie 0.01 probability of overtopping). This topic is discussed further in reference 10.

Simplified assessment

4.3.21 For very small schemes, or preliminary studies, properties can be grouped for the purposes of damage estimation. Estimates can then be used of the likely average depth of flooding in each group of properties. This procedure should be applied for each of a minimum of 2 flood events above the threshold flood. The average loss for a residential property at each of those depths, multiplied by the number of properties in each group, can then be used to derive the loss-probability curve. For a very basic assessment, it is possible to consider the average benefit of protecting a residential property. An analysis of the benefits of flood prevention schemes suggests an average annual flood damage of between £3,000 and £8,000 per property in a moderately vulnerable area (reference 1).

Use of software packages

4.3.22 A variety of dedicated software packages for flood alleviation benefit assessments is available and spreadsheets can also be used (reference 1), particularly for the smaller and simpler schemes. In addition to reducing simple arithmetic errors, a virtue of these systems is in archiving data and recording the bases of different assessments. They also allow rapid recalculations, as the assessment is progressively refined during the course of the project.

Validating the inputs

4.3.23 Mistakes can easily arise when entering large quantities of data. It is therefore important to check for obvious errors in the input files. In particular, discrepancies in property floor levels, areas and grid references may lead to large errors. Scatter plots and frequency distributions are quick and simple ways of picking up extreme values, and hence checking for potential input errors. Some programs can be set only to accept locally valid ranges of values.

Checking results

4.3.24 The results of a program also need to be validated. Particular care should be taken where simplified loss models are effectively required to act as overland flow models, converting flood level data for points along the river channel to flood depths for particular properties. Ideally, an overland flow model would be used to generate these depths, to take account of local topography. This is often not possible and, consequently, engineering judgment combined with local knowledge should be used to make reasonable estimates of the correct flood depths.