A global multi-hazard risk analysis of road and railway infrastructure assets

Keywords: Environmental impact, natural hazards, transport infrastructure, road, rail

This research was funded by a grant from the UK Engineering and Physical Science Research Council to the ITRC

Transport infrastructure in any country of the world is irrevocably exposed to one or more types of natural hazard. 

While many developed countries like the UK have a grasp of their own transportation networks and their exposure to natural hazards under current climatic conditions, there has long been a need to apply a globally consistent framework to understand the big picture and global risks the transportation sector faces – particularly as we approach uncertain future climate conditions.

The study demonstrates the potential for conducting infrastructure risk analysis at a high spatial resolution on a global scale.

The multi-hazard risk analysis presents the first global estimates of multi-hazard exposure and risk to road and rail infrastructure. It finds that around 27% of all global road and rail assets are exposed to at least one hazard and around 7.5% of assets exposed to a 1/100 year flood event.

The findings illustrate that global damage to road and rail infrastructure due to natural disasters could result in annual costs on average of approximately $14.6 billion, and that countries can save billions by systematically considering risks and protecting transport infrastructure assets from the planning stage.

The findings show that it is essential for countries to improve their transport planning by including risk information in their assessments, particularly around flooding exposure with over 73% of expected annual damages arising from surface and river flooding, and a further 15% of annual damages attributed to coastal flooding.

As well as offering cost savings from improved transport planning, it supports the United Nations’ Sustainable Development Goal (SDG) 9 which calls for increased access to sustainable transport infrastructure in low and middle income countries, and SDG 11: make cities and human settlements inclusive, safe, and sustainable.

Research question and scope   

The research sought to further scientific research into the global impacts of natural disasters on infrastructure networks, both through direct infrastructure damage and indirect impacts on users and supply chains.

The researchers believe this is the first global study that addresses damaged networked infrastructure at the asset level, such as individual road segments or bridge structures.

The majority of previous research focuses on single hazards in isolation whilst this research comprehensively maps multiple natural disaster risks and their impacts on transport infrastructure, namely road and railway assets. The most frequently recorded and costliest disasters are included in the research: tropical cyclones (wind speed only), earthquakes, surface flooding, river flooding, and coastal flooding.


To carry out the modelling, the researchers used state-of-the art global hazard mapping, combined with innovative analysis of approximately 50 million km of transport network data included in OpenStreetMap, and assumptions about the fragility and (re)construction of transport infrastructure derived from a variety of sources.

Due to the large size of all data sources (both in storage and in information), the researchers  split the analysis over 46,566 regions based on the GADM administrative level 1 and 2 datasets. By using parallel and cloud computing, runtimes were brought down to a reasonable time-scale, allowing for a global risk analysis with this level of detail. 

This study includes earthquakes, tropical cyclones, and surface, river and coastal flooding. For the exposure analysis, hazard data is reclassified into hazard intensity bands.

All road and railway data are based on open access data from OSM. 

Infrastructure damages are estimated using a variety of sources of replacement cost data and fragility curves. 

We performed a cost-benefit analysis on each road segment. The CBA estimates the benefit-cost ratio (BCR) of upgrading the each road by spending 2% of the road’s value on barriers and better drainage (assuming roads are upgraded at the end of their lifetime).


Global exposure of transport infrastructure assets is presented in across 46,566 regions.

Building in additional resilience such as improved road design for better drainage and including flood barriers at planning stage, at an additional cost of around 2% of the road value, could result in financial savings of 60% of all the roads that are exposed to at least one flood event or more; and of over 80% of the primary and secondary roads flooded on average every year in upper middle income and tertiary roads in lower and middle income countries.

The study shows that although absolute damages are found highest in US, Japan and Europe, Small Islands Developing States (SIDS) and lower income countries are hit hardest in terms of GDP.

Results also show:

  • The global EAD to transport infrastructure assets 
  • Exposure to risk per hazard and per intensity band for the four income groups
  • The 20 countries in which the transport infrastructure is most affected by natural hazards 
  • the 20 countries that have the highest multi-hazard EAD in absolute terms
  • The countries which are particularly vulnerable. Myanmar, for instance, is experiencing one of the highest absolute levels of risk to its transport infrastructure, but also the highest risk as a percentage of GDP and one of the highest per kilometre of road.

Key findings

Simple ideas such as prioritising those most exposed assets to flooding over blanket increases in flood protection standards across a network can be a major risk reduction.

Modelling has shown that as little as 2% of total build costs spent on extra flood protection can result in 60% reduction in flood damages from 1-in-100 year flooding events.

Global risk estimates of all floods combined can be reduced up to 42% when upgrading the roads to design standards (Supplementary Table 7) that halves the annual probability of flooding (i.e., upgrading the design standard to withstand a flood with 1/100 return period instead of a flood with 1/50 return period). 

In absolute terms, the largest reduction in damages can be achieved by upgrading the design standards for surface flooding, as this type of flood is dominant in all income groups.

In relative terms, the largest gains can be made for coastal flooding in low income countries, with a reduction of up to 80% in risk. This large gain is primarily due to higher inundation depths for coastal flooding compared to surface and river flooding.


While undertaken at the global scale, the methodology and code used to do the analysis is openly available and free for anyone to use such that the study can be carried out on a far more granular scale for individual countries or regions. Ultimately this becomes an essential tool for identifying and prioritising assets most at risk to wider transportation networks and when conducting initial planning studies.

The mapping and analysis carried out for this paper could be carried out on a more granular scale for individual governments or countries. This methodology can be used to identify where the single points of highest risk and vulnerability are in their transport infrastructure system. It can also identify risks to take into account when planning new road and rail systems.

The analysis can be used to help researchers identify the potential benefit-cost ratio of upgrading roads to reduce the risk of flooding to road assets.

It is crucial that countries, when exposed to natural hazards, improve transport planning by systematically including risk information and improving the protection of their most vulnerable and critical assets. There is a need for better risk information to avoid spending more on all assets, but being able to spatially target improvements. The economic and social benefits to be gained from doing so would go well beyond direct infrastructure damage.

What’s next

The authors are currently researching other impacts on infrastructure such as the impact and cost of disruption to supply chains in global transport networks, and the exposure and risk of global power infrastructure to natural disasters.


ITRC-Mistral in conjunction with other researchers from the University of Oxford, the World Bank, and the European Commission Joint Research Centre.


Nature Communications, ‘A global multi-hazard risk analysis of road and railway infrastructure assets’ led by Dr Elco Koks of ITRC-Mistral in conjunction with other researchers from the University of Oxford, the World Bank, and the European Commission Joint Research Centre.


Elco Koks

University of Oxford Read more