This report describes the work carried out on the NERC-funded project ‘Assessing the risk to the coastal and rural road network in Scotland due to the effects of storms and extreme rainfall events’ Grant Reference NE/N013034/1. The project is part of NERC’s Environmental Risks to Infrastructure Innovation Programme, which aims to utilise existing environmental research to meet the needs of industry. The project was carried out by the University of Dundee and TRL, in conjunction with Transport Scotland, who supported the project by providing resources in-kind. The aim of the project was to develop a methodology for assessing the risk to infrastructure from coastal hazards. The risk assessment methodology was based on a similar approach developed by TRL to assess the hazards and risks associated with landslide events. An assessment of the economic costs of the event was undertaken and based on a methodology also developed by TRL. The methodology was developed using a case study site selected in conjunction with Transport Scotland. The site selected was the section of the A78 that runs along the west coast of Scotland between Skelmorlie and Largs. This location has frequently experienced flooding in the past, often leading to the complete closure of the road and a long diversion for travellers. Obtaining data on past events associated with natural hazards was challenging as the details of incidents and their impacts are often not recorded. However, information on a coastal flooding event on 3 January 2014 was available and this was used as a representation of a typical flood scenario for developing the methodology. A visit to the case study site was carried out in order to take and verify measurements of the level of the road and the distance from the road to the high tide and to determine the characteristics of the road, shoreline and surrounding area. Evaluation of this information confirmed the presence of many of the factors associated with susceptibility to coastal flooding. Much of this section of the A78 is on low-lying land very close to the high water mark and has limited intertidal width. There is also a lack of natural wave breaking features such as sand dunes. Sea level measurements from 1990 onwards were obtained from a nearby tide gauge at Millport. A Flood Potential Value (FPV) was devised from the gauge data comprising of the astronomical tide plus two times the difference between the measured sea level and predicted astronomical tide (residual). The residual is the surge associated with the meteorological conditions; it is multiplied by two to account for wave height. The benchmark flood event was used to define a FPV threshold above which flooding at the case study site is highly likely to occur. It was noted that the FPV data correlates with a wellknown index of weather and climate change for the North Atlantic region known as the North Atlantic Oscillation (NAO). Changes in monthly NAO Index values based on air pressure data provide information on changes in the position of the winter North Atlantic storm track and, by inference, of the mi-latitude jet stream. The project results show a correlation between periods of winter storminess along the Ayrshire coast with periods when the NAO Index was positive. It is concluded that this relationship may form the basis for flood preparedness in the future. This particular conclusion may have a wider relevance to the evaluation of coastal flood risk for the country as a whole. The UK Climate projections (UKCP09) sea level changes estimates show an increase in the relative sea level at the case study site (estimated to be between 0.2 — 0.39m by 2100 depending on emissions (central estimate)). It is uncertain from current climate models if the frequency of storms impacting on the British Isles will change, but an increase in sea level alone will result in the FPV threshold being exceeded more often. The number of times the threshold is likely to be exceeded in 2025, 2050 and 2100 was calculated using the central estimate of sea level change for low, medium and high emissions. This gave an increase in the number of times the FPV is exceeded from the current average of 1.1 times per annum to 2.2 times per annum in 2100 under medium emissions. This represents a significant increase in the number of flood events experienced. Both the direct and direct consequential economic impacts of the benchmark flood event were estimated. The direct economic impact comprised of the costs of clear-up and repairs to the infrastructure. The direct consequential impacts are the costs that result from and vehicle incidents during the event and the modelled costs of user delay, carbon emissions and associated changes to the incidence of traffic accidents. It should be noted this is not the full cost of the event; indirect costs from the longer-term impacts on local businesses and tourism (etc.) were not included in this study. These indirect consequential impacts are difficult and costly to measure and, when the associated costs are not trivial, the resulting costs data can be open to wide-ranging interpretations The cost of repair and clear-up was obtained from the operating company, Scotland TranServ. They also provided traffic flow data which was used in the calculation of the direct consequential costs. During the benchmark event a storm surge led to a series of closures as the situation escalated and then receded. There was also some damage to the sea defences and adjacent road structure, which prevented complete re-opening of the road until it was repaired. The QUADRO model, developed for determining roadworks delay, carbon and accident costs (the latter as a result of the roadworks) was used to model the traffic costs for these different types of closures and traffic restrictions. Provision is made within the methodology to combine these with incident accident costs to give the total direct consequential cost per event, although in the case study detailed in this report there were no known incident accident costs. Relating these results to the frequency of event per year provided a cost per annum for today, 2025, 2050 and 2100. These costs were based on current vehicles and traffic levels, however in reality technology and travel demand could change considerably over the next 84 years. In order to explore how future changes could influence the risk, the model was run using traffic levels for 2100 based on current traffic trends (i.e. an increase of 2.4 times current traffic levels). This increased the risk considerably illustrating a significant sensitivity for traffic levels. The results from this type of risk assessment can help to inform the adaptation actions of infrastructure owners, providing them with information on the scale of the problem and how this is likely to change due to climate change. This can provide the evidence required to prioritise resources, plan budgets and form a business case for action. In addition to the methodology developed and the results produced for the A78 case study site, this project identified a number of actions that would for improve the robustness of this type of calculation and therefore the information available to infrastructure owners. These include improvements such as the inclusion of flood risk factors in asset inspections, establishing a national geohazard database to record the details of incidents and their consequences and developing a national hazard coastal map which includes morphology. This project was a small scoping study based on one case study site, so a further recommendation is for a more extensive study to be carried out covering different types of coastline and infrastructure. The methodology developed is considered to be broadly applicable also to the rest of the UK, and potentially beyond, and to other types of infrastructure in the coastal hinterland, such as railways. (Author/publisher)
Samenvatting