The Scottish Government

« Previous | Contents | Next »

Listen

6 Findings and conclusions

6.1 Reporting flood warnings

At present information on flood warnings and watches are provided as lists, flood warning area and status and this assumes that the consequences of flooding are similar in each area. An alternative indicator of the area likely to be at risk would be in a map or graphical form that illustrates properties at risk or potentially at risk properties. This information will be available within the Second Generation Flood Risk Maps currently being prepared by SEPA and will provide the agencies involved in flood warnings with an assessment of the hazard presented by the flood risk.

6.2 Flood reports

In terms of gauging the impact of flood risk, such a map defining the flood extents would allow an assessment of costs to property and other assets. An estimate of flood depth at a property allows a straightforward valuation of damage. Other costs can be reported separately, i.e. clean up costs or dealing with road diversions or mobilising assistance to residents if recorded.

After the flood in August 2004, Scottish Executive sought local authorities to provide information on the flooding in their area. The information submitted is kept in a file and varies in format. In order to improve the establishment of future flooding information the completion of a standard proforma by a local authority would allow timely and consistent data collection. This could be easily completed, perhaps on line by an agency to provide data and could be included in any biennial return. An example for Menstrie has been included in Appendix E.

SEPA have recorded a number of historical outlines. Scottish Water collects data reported to its customer careline and therefore often has the best dataset in urban areas; however it is property specific and can be considered sensitive. Fire Brigade, Police and Local Authorities also record flooding incidences. This information is not collated centrally and therefore this leaves an important gap in our understanding of flood risk, particularly in economically important areas. The costs associated with urban and sewer flooding are significant and, although often considered in isolation by the relevant authorities, the costs to the nation need to be aggregated to determine an overall cost.

It is recognised that it is difficult to mobilise staff to record areas of flooding in urban areas before wrack marks or water marks are removed in clean up operations. Therefore it is necessary to know which authority will be responsible for collecting this data and how it will be resourced in advance.

6.3 Flood severity

Only limited data was made available within the timescale of the project and some of the estimates are made on a very limited data set. August is traditionally a high precipitation month although extensive flooding is less common due largely to the river levels being fairly low and higher rates of evapotranspiration associated with summer conditions. August 2004 was judged to be exceptionally wet and report rainfall and river flows well above the monthly average.

Whilst the peak flows in the larger rivers in August 2004 are not large when assessed against the annual flood series they would appear to be locally significant for August. The rainfall is similarly distributed. This would also suggest that for smaller catchment and the areas affected by landslides where there are no gauges either within the catchment that the event could be more severe.

In terms of quickly assessing the flood severity the following steps could be taken:-

• The hydrological summary is rapidly available after a flood event and could be used in assessing return periods and event's severity. Data is summarised from SEPA and the Met office gauges and compared using standard methods and hydrometric areas. SEPA use the hydrological summary to ascertain return periods in publications such as their bathing water report ¹⁹.
• The primary hydrometric data can be analysed in detail, this could be undertaken by SEPA and provided to the Executive or undertaken by a third party on supply of the long term records. There would appear to be resource issues in providing the original data. Some data in key areas affected by the August floods took 4 months to be made available. If this analysis is likely to be a regular requirement then resourcing issues may need to be addressed. This could be achieved by increased staff or placing data in the public domain through the web. The data is currently held in a number of databases and some of the analysis required could be via standard query within these databases.
• A software tool to provide information on flood severity could readily be written.

6.4 Economic costs

The overall damages have been estimated at between £7.2 million and £31 million. With the mid estimate at £18 million, this is equivalent to uncertainty (around the mid estimate) of between 60% and 70%. The non-money impacts are expected to be small and are unlikely to significantly affect the damage estimates, such that it may be considered that the 'true' damages fall within the above range. To improve these cost estimates, much more quantitative data is required on the extent and effects of flooding. In particular, key information that should be recorded during and after flood events includes;

• number of residential properties flooded (some information was available, but this was not complete);
• number and type of commercial properties flooded (this is a key source of uncertainty and more research may also be required into the damages that are experienced by commercial properties);
• number of weeks that people have to move out of their property while it is being repaired (no information was available on this);
• area of land (agricultural and environmental) affected by landslides and/or flooding (no quantitative information was available on this);
• costs of emergency repairs to flood defence structures and number/type of flood events that incur this expense (no information was available on this making the estimates highly uncertain);
• length of diversions following landslides/road closures due to flooding (information was available on the larger diversions, but not on smaller events which could result in an under-estimation of total costs);
• time over which disruption to train services was experienced, particularly for major lines; and
• social impacts of flooding, some of which may not become apparent until some time after the flood events. This will assist in determining whether the non-money impacts are likely to be small compared to the money impacts or whether they are larger than predicted.

There are also a number of damage cost estimates that are based on data for England and/or Wales. It is therefore unclear whether this damage data provides a definitive indication of the costs faced in Scotland, suggesting that some work may be required to either obtain specific cost estimates for Scotland. The cost estimates are however used are those that are routinely used in the assessment of flood defence schemes in Scotland. It would be possible to compare the figures with those in the Dundee Flood Loss tables ²⁰ or to check whether the estimates used in this study are broadly applicable or not.

6.5 Environmental costs

Some of these can be quantified in the economic costs. There are others that as discussed in our environmental standards that are less certain. SEPA suggest that fluvial water quality will decrease in a wet year due to disturbance and increased diffuse run off while estuarine water quality may increase due to dilution. The increased categorisation of water bodies underway as part of the Water Framework Directive may provide measure that allow a comparison and a value to be attached to this type of impact.

6.6 Recommendation for further study

The costs indicate that flooding in August contributed a significant cost to the economy. In addition publication of this type of report may prompt further information to be brought forward. Other bodies are still carrying out their own investigations into the flooding including:

• severity of flooding in Glen Ogle - SEPA and Universities
• Scottish Water -costs of sewer flooding
• Local authorities in terms of new works

It is likely that the costs and understanding of the August 2004 event could be refined.

The costs of flooding in August were estimated on the basis of data collected for this study and in many cases this inadequacy has been highlighted.

In order to improve and enhance the collection of flood event data, the key sources of information on flooding are summarised below Table 6-1 with an indication of the data type that each body may hold. In establishing accurate cost for flooding it is necessary to identify the number and type of properties affected. A number of organisations hold this information, and in order to develop a clear picture it is necessary to pool and compare this information. At present there is not one body with an over arching responsibility for undertaking this task or with legislative powers to do so. Some organisations have expressed concern on data protection issues in terms of sharing data on properties affected. The information is also held in a number of forms.

6.7 Coordination of event analysis

At present there is no single body with responsibility for flooding and as such there is not a body that collates all of the information relating to costs for flooding in Scotland. To make a complete assessment of costs it would be necessary to firstly collate all the information and then analyse the data. This would require consistent collection of data relating to flooding by relevant authorities and the subsequent collation either by an external party or Scottish Executive to determine the full extent and cost of a flood. The main constraint on this is data protection as information is useful for insurers and for property values.

Completing a detailed exercise for a number of floods would allow an assessment of the performance of the standard methodologies used in this report and their strengths and weaknesses for future application. This would also provide guidance on how coordination of data collection and its subsequent analysis could be undertaken.

« Previous | Contents | Next »