BT16 Rail network benefiting from fluvial flood protection

Transport supports many different social and economic functions. In 2013/14, 86.3 million passenger journeys were made by rail in Scotland - 16% of all public transport journeys.  Rail patronage has increased by 35% since 2004/05 and constitutes a growing share of public transport journeys.  In contrast, there has been a decline of 8.2% in the amount of freight (tonnes) lifted by rail between 2002/03 and 2012/13.  The modal share of freight (in tonne-km) carried by rail in Scotland in 2010 was 7% (Transport Scotland, 2014).

Flooding of rail infrastructure can cause disruption to rail transport with knock-on consequences for these functions – e.g. preventing or delaying people from reaching work, delaying rail freight. Climate change predictions suggest that flooding of rail infrastructure will become more extensive and more frequent. This indicator will provide insight into action taken to protect the rail network.

Related indicators:

BT8 Railway network at risk of flooding

BT9 Disruption risk to railway services as a result of flooding

BT12 Flood events affecting the railway network

Legislation has been a key driver of historic flood prevention activity.  The Flood Prevention (Scotland) Act 1961 provides powers to enable Councils in Scotland to take measures for the prevention or mitigation of flooding of non-agricultural land in their areas (HMSO, 1961). Some 72 flood prevention schemes were funded and delivered in Scotland under this legislation (JBA Consulting, 2012). The objectives of these flood prevention schemes focus on the protection of commercial, industrial or residential properties in Scotland (Bassett et al, 2007). As such, any protection of rail infrastructure appears to be a secondary consideration. The nature of flooding is such that most flood prevention schemes provide for the management rather than prevention of flooding (JBA Consulting, 2012). In effect, most schemes built under the 1961 Act are ‘hard engineered’ interventions (e.g. embankments) located in flood risk areas designed to help prevent localised flooding where the consequences of a flood would be severe.

Network Rail’s Scotland Route Weather Resilience and Climate Change Adaptation Plans include a range of ongoing and proposed actions to help address flooding impacts to the rail network, including impacts that are expected to worsen or become more frequent as a result of climate change (Network Rail, 2014). This includes actions at specific sites to address known flooding issues as well as more general measures to reduce the vulnerability of the rail network to current and possible future flood risks including an asset strategy to upgrade small diameter culverts (e.g. on the West Highland Line) and various measures to make track assets more resilient to flooding (ibid).  

The 2009 Flood Risk Management (Scotland) Act introduced a more sustainable approach to flood risk management (FRM) suited to the impacts of a changing climate (Scottish Government, 2015). In particular, the 2009 Act establishes a framework for coordination and cooperation between FRM stakeholders, helping to pave the way for catchment scale approaches including the use of natural flood management (e.g. restoring watercourses to a more natural form by reducing morphological pressures, planting vegetation and managing slopes to reduce runoff at source) to help restore more favourable runoff patterns and reduce flood risk (Scottish Government, 2011). Although there will always be a requirement for more traditional ‘hard engineered’ flood prevention schemes (e.g. embankments) in high risk areas, the 2009 Act establishes a statutory basis for considering wider catchment or land based approaches that may be more resilient in the face of climate change (ibid). In consequence, it may be the case that the location, scale and type of flood risk alleviation measures adopted in the future changes to reflect the 2009 Act’s more integrated, catchment scale approach. For example the use of land management measures in rural upper catchments could reduce the need for flood embankments in urban areas lower down the catchment.

The National Flood Risk Assessment (NFRA) carried out under the 2009 Act identifies a number of Potentially Vulnerable Areas (PVAs) across Scotland (SEPA, 2011). The PVAs highlight areas that are particularly vulnerable to flooding based on a range of different receptors as well as proxies for receptor value and vulnerability (SEPA, undated). In effect, the more receptors exposed to flooding (and the more valuable and vulnerable they are) the greater the overall risk, whereby risk is a function of event likelihood and consequence (Scottish Government, 2011). Transport (roads, railways and airports) are included as receptors in the NFRA (SEPA, undated). The PVAs are used to focus FRM effort to the locations where actions will deliver the greatest benefit by reducing overall flood risk. As a result, FRM action under the 2009 Act is likely to take a more strategic, targeted approach. Also, the criteria used in the NFRA (e.g. rurality provides a proxy for vulnerability) means that PVAs are spread variously around the country and not just concentrated on densely populated urban areas (though most PVAs do cluster around the central belt).

Thus it is anticipated that the type, scale, location and strategic targeting of flood prevention measures will change in the future, given the focus on risk based, catchment scale approaches. As the contents of the forthcoming Flood Risk Management Strategies (FRMS) and Local Flood Risk Management Plans (LFRMP) are not yet known it is not yet possible to determine how the rail network might benefit. However, given the nature of the PVA approach, it seems plausible that the proportion of the network benefitting from flood prevention measures will increase. 

The design of flood prevention schemes is constantly being improved through the use of design flows or floods that account for climate change and through more detailed hydraulic models (Bassett et al, 2007). As a result, future schemes are likely to be better designed to cope with the demands of climate change.

There are several key limitations to the assessment as summarised below:

1. The SFDAD aims to be fully comprehensive although there are some gaps – e.g. in terms of historic (1961 onwards) flood prevention scheme data that has been destroyed, lost data as a result of restructuring of Local Authority boundaries (e.g. as a result of the 1994 Local Government Act) and the scope of the SFDAD which is focussed on schemes constructed under the Flood Prevention (Scotland) Act 1961 (Angus Pettit – JBA Consulting Principal Analyst, personal communication, June 4, 2015). Furthermore, Bassett et al (2007) highlight inconsistencies between flood prevention scheme data held centrally (by the then Scottish Executive) and locally by Local Authorities – e.g. some Local Authorities were unaware of schemes within their jurisdictional area and others held information on schemes that were not held centrally. It is not clear if (or to what degree) these data management issues have been resolved through the SFDAD. Any under representation of flood defence schemes within the SFDAD would affect the comprehensiveness of this assessment.
2. The intention is for the SFDAD to be kept up to date and for new schemes to be added when data is made available by Local Authorities (Angus Pettit – JBA Consulting Principal Analyst, personal communication, June 4, 2015). However it is not clear if or how the SFDAD is kept current in terms of scheme maintenance, condition and upgrade, all of which could influence the SOP afforded by the scheme. For example, Bassett el al (2007) highlights a number of projects and activities that were undertaken even during the course of the SFDAD project (e.g. repairs, maintenance, updates to surveys or modelling). It may be the case that interventions such as these are commonplace and associated alterations to schemes would affect the accuracy of the assessment.
3. The SOP afforded by schemes covered within the SFDAD is based on design flows or floods for a given return period at the time of construction. Bassett et al (2007) highlight how in the majority of cases, updated flow or flood estimates (e.g. those undertaken as part of the SFDAD project) are higher than those used in the original scheme design. This could be caused by a number of factors including climate change (e.g. higher rainfall and associated flows means that higher magnitude flood events are becoming more frequent), land use change and hydraulic changes to channel, banks and floodplain upstream of the scheme. Bassett et al (2007 p.64) conclude that “the results of this study indicate that few schemes currently protect to the designed standard”. This could affect the accuracy of the assessment – e.g. if schemes are selected for assessment the basis of the SOP field in the SFDAD data (see Assessment Step No.2 in Table 4) yet they are not providing 200 year SOP for current climatic and hydrological conditions.
4. This indicator only considers the benefits of fluvial flood protection measures. There is no consideration of rail network design for mitigation of pluvial or drainage flooding^[1] or the benefit afforded to the rail network by coastal flood prevention schemes. Furthermore, it is anticipated that a number of schemes designed to manage coastal erosion or provide coastal protection under the Coastal Protection Act 1949 will also provide an element of flood protection (JBA Consulting, 2012). The flood risk alleviation benefit of these schemes is currently not considered. As such, it is likely that the extent of the rail network benefitting from flood prevention measures is greater than stated here.
5. Major railways located in the floodplain are often raised above the ground surface on embankments. The difference in elevation afforded by these embankments is not always identified in flood modelling and mapping (Thornes et al, 2012). As such it may be the case that the flood protection (from SFDAD schemes) to rail infrastructure is over-estimated – i.e. where the embankment itself would raise the railway above the protected area.
6. Network Rail’s Network Links Layer which was used in this analysis includes double sections of track at many locations as well as railway siding etc.  It should be noted that the total length of track used in this analysis does not correspond with the length of track published by Transport Scotland (2014) which counts one kilometre of single or double-track as one kilometre of route length.
7. Due to resource constraints assessment was only undertaken for 0.5% probability (1:200 year return period) modelled flood events.  These are low probability events and are located at the more severe end of the flood event spectrum.  It should also be noted that the magnitude of a 1:200 year event is based on historic data and may be greater in the future as a result of climate change than represented in this analysis.

ClimateXChange (2016) Adaptation to Climate Change: Context and Overview for Transport Infrastructure Indicators. Available online at our Indicators and trends pages

Analysis and development was undertaken by Dr Neil Ferguson (University of Strathclyde) and Dr Peter Phillips (Collingwood Environmental Planning Limited).

Katherine Beckmann, Heriot-Watt University / CXC contributed to this indicator.

SEPA and Network Rail provided the bulk of the data underpinning this assessment (SFDAD polygons and rail network polyline respectively).