The Scottish Government

« Previous | Contents | Next »

Listen

Natural Flood Storage and Extreme Flood Events Final Report

1 CONTEXT

1.1 Background

Flooding is an important process in the natural environment and cannot be entirely prevented. The natural characteristics of a catchment, together with man-made influences, determine the generation, duration and extent of an out-of-bank flood event. Over many centuries mankind has influenced the use and extent of natural floodplains in Scotland. This has generally reduced the connectivity of the river network to historical washlands, flood meadows and wetland areas that once featured extensively on floodplains and can potentially hold water back during a flood event. Artificial land drainage (including the infilling of small lochs and wetlands), followed by land use and land management change towards more intensive agricultural production, or settlements and engineered flood protection structures, further restricted the amount of floodplain available to convey and store flood waters. However, other floodplain areas still exist (generally under rural, agricultural or conservation management regimes) that could potentially be utilised to help reduce the flood risk downstream. Natural flood storage is now being considered to be in principle a useful complement to conventional flood defences that allows for more environmentally sustainable flood management.

This research project was established to investigate and quantify the potential of natural flood storage on Scottish rivers. One of the motivations is to plan for increases in extreme rainfall events and increased river flow. Current climate change predictions suggest that substantial areas of Scotland (particularly eastern Scotland) will experience a greater frequency of flooding during the winter, due to greater and more intense rainfall (Price and McKenna, 2002). Summer flooding is, however, predicted to become less common, although changes to summer thunderstorm pattern, intensity and potential flood impact have not been quantified. Floods that are considered extreme today could become more common in the future. Current estimates suggest that the inland floodplains in Scotland cover an area of around 2,950km ², putting about 77,100 residential or commercial properties at risk from fluvial flooding (Werritty et al, 2002).

The Water Framework Directive (WFD) (2000/60/EC) represents the most significant piece of European water legislation for decades. The WFD, transposed into Scottish Law by the Water Environment and Water Services Act, 2003 (WEWS Act), requires all Member States to achieve prescribed ecological standards for all waters and to "contribute to mitigating the effects of floods". The WEWS Act specifically requires Scottish Ministers, the Scottish Environment Protection Agency (SEPA) and other responsible authorities to promote sustainable flood management. Sustainable flood management requires a catchment-wide approach to be taken that utilises and enhances natural systems within the floodplain. There will always remain a place for hard defences in future flood management because in certain circumstances there may be no alternative. However, these may be used together with appropriate soft engineering techniques.

Recognition of the role that flood storage has to play in sustainable flood management within the UK has been emphasised in a number of recent publications, such as 'Learning to live with rivers' (ICE, 2001), 'Turning the tide on flooding' (WWF Scotland, 2002), 'Wetlands, land use change and flood management' (EN/EA/Defra/FC, 2003), 'Go with the flow' (RSPB Scotland, 2004) and the report of the COSLA Flooding Task Group (COSLA, 2003). In addition, the role that farming can play in sustainable flood management was highlighted in the report of the Policy Commission on the Future of Farming and Food - the 'Curry Report' (Cabinet Office, 2002).

1.2 Definitions

The concept of 'natural flood storage' is potentially confusing and should be avoided. Instead, it is often useful to consider natural flood attenuation, which may include some storage effects as well as floodplain conveyance. Natural flood attenuation is built into any flood event by definition. Making additional use of areas that flood naturally has to involve some form of engineering or active flood management, such as maintaining flood banks or adding roughness elements to the floodplain.

For the purposes of this project the following definitions have been used:

Natural flood attenuation - the overall impact of the floodplain to change the shape of the hydrograph (reduce flood peak and increase flood duration) during out-of-bank events due to a combination of storage and resistance to flow.

Flood event - a combination of hydrological factors that leads to a downstream flood flow of a specific probability (or return period).

1.3 Aims and Objectives

The overall aim of this project was to assess the applicability and use of 'natural flood storage' available within a river system and how this could be enhanced to reduce the risk of flooding to settlements in the catchment. The specific objective of the project was to develop a robust methodology to assess:

1. The probability (or return period) of flood events that could be contained by the existing natural flood storage within case study catchments and mitigate flooding downstream.

2. The additional storage capacity, above this natural capacity, that would be required to prevent a 1 in 100 year and a 1 in 200 year event in a catchment from causing flooding downstream.

3. The location and feasibility of providing (where available), such additional storage capacity.

Objective 1 is conceptually difficult because 'natural' storage or attenuation is not fixed, but is instead a function from the extent and depth of flow on the floodplain. Hence, for any flood flow downstream, the 'natural flood storage' is contained within the upstream area and volume of flooding, whatever that may be.

Objective 2 can be investigated by comparing hydrographs for the 100 year or 200 year event with the hydrograph or peak flow that represents a threshold of flooding at some downstream risk location. For the purposes of this project, it has been assumed that the 200 year event could provide one possible definition of areas on the floodplain that could provide 'natural' storage. The question is then whether the required volume of storage can be found within the area flooded during a 200-year event. (The assumption of the 200-year event is arbitrary; a larger event could be a better way to define the natural floodplain and could easily be used. It was thought that a 200 year event would provide a reasonable choice to define a large natural flood extent without extending far into built up areas).

Objective 3 was addressed by undertaking a geographical analysis of the land cover and other natural or man-made assets for the various flood scenarios to determine the practical, economic and environmental implications of inundating the floodplain areas as a flood mitigation measure.

The project team was directed by the Scottish Executive to work with four case study rivers that had been used during previous research studies on flooding in Scotland and for which various appropriate datasets and/or river models (covering different spatial coverages) already existed. The four case study rivers were:

1. Clyde, draining through Glasgow

2. White Cart, a tributary of the Clyde, draining through southern Glasgow

3. Tay, draining through Perth

4. South Esk, draining through Brechin

The case study rivers represent a range of catchment characteristics (e.g. area, topography, land cover, rainfall), river modification and regulation regimes, and degree of floodplain development (Figure 1-1 and Table 1-1). These characteristics will influence the generation of runoff from the land and the conveyance of flood water down the catchment, whether in the channels, on the floodplain or temporarily stored in reservoirs/lochs.

Source: Derived from CEH Land Cover Map 2000

1.4 Datasets and information gathering

The project required the acquisition, review and use of numerous datasets and other information on the four case study rivers, which were obtained from a variety of organisations, both in Scotland and England. The datasets that were acquired fell into a number of categories:

1. Topographic data - e.g. digital elevation model (DEM) such as LiDAR, NEXTMap or OS Profile

2. Hydrological data - e.g. river network, open water bodies

3. River model data - e.g. routing or hydrodynamic models

4. Land Cover data - Centre for Ecology and Hydrology (CEH) Land Cover Map 2000 (LCM2000)

5. Environmental designations and other environmental data - e.g. Sites of Special Scientific Interest (SSSIs), Scheduled Ancient Monuments (SAMs), Listed Buildings (LBs), River Habitat Survey (RHS)

6. Floodplain asset data - e.g. electricity, transport infrastructure

In addition to the datasets that were acquired, contact was made with key stakeholders in the case study river catchments who would have an interest in the project. Some of these would also be part of the formal consultation process in the event of any potential new flood storage area being further investigated as a future flood management option in a particular catchment. The organisations that were contacted for the provision of data or other information were:

• Scottish Executive
• Scottish Environment Protection Agency (SEPA)
• Scottish Natural Heritage (SNH)
• Historic Scotland (HS)
• Macaulay Land Use Research Institute (MLURI)
• Glasgow City Council
• Angus Council
• Perth and Kinross Council
• Environment Agency of England and Wales (EA)
• Scottish Water
• Scottish and Southern Energy Group
• British Telecom

Not all the organisations contacted above have provided datasets or information to the project. Some of the organisations requested payment for the supply or lease of the datasets. In these instances the datasets were not acquired for the project, but their availability and use are noted where appropriate in this report.

1.5 Digital elevation model (DEM)

The assessment of natural flood storage through the application of flood inundation models required the use of good quality and accurate topographic elevation data. Ideally, this would mean that the generic methodology was based on a single standard DEM which was available for the whole of Scotland, rather than making use of different DEMs (e.g. LiDAR or OS Profile) for different areas depending on the coverage. During summer 2004 it became known that the NEXTMap DEM that had previously only been available for southern Scotland was expected to become available for northern Scotland as well. As a result the generic assessment methodology has been directed to make use of the NEXTMap DEM. The new NEXTMap DEM for northern Scotland became available to the project at the end of September 2004.

The suppliers of the NEXTMap DEM in Scotland, Getmapping, quote the vertical accuracy as +/-1.5m root mean square error (where the height from which the aircraft gathered the radar data was 30,000 feet. The horizontal accuracy is quoted as being +/-2.5m on slopes less than 20 degrees, which will include all floodplain areas. The grid resolution is 5m x 5m.

« Previous | Contents | Next »