Counting the cost of last winter’s flooding: Evidence from disruptions to the road network

[Post co-authored by Paul Kilgarriff and Tom McDermott]

This time last year Ireland was in the middle of its wettest winter on record [PDF]. A series of Atlantic Storms battered the country, beginning with Storm Desmond in early December, followed by Eva and Frank. Rainfall levels in some areas were up to 300% of normal levels. Extensive flooding around the country caused widespread damage – hundreds of homes and businesses were flooded, and thousands more were cut off by flood waters. In many places the floods did not recede until well into the new year.

Various impacts of the flooding are detailed in the recent report by the National Directorate for Fire and Emergency Management (NDFEM)[PDF]. Almost €1.8million in humanitarian assistance was paid out to affected households; close to €1m to farmers; local authorities received special funding of €18m for clean-up costs; while damage to the road network was estimated at over €100m. Aside from damages, the flooding also caused substantial disruptions to everyday life — 350,000 customers suffered disruptions to electricity supply, and 23,000 households were placed on boil water notices. The flooding also resulted in substantial travel disruptions – in particular as a result of flooding on the road network.

While the costs of direct damages (destruction of assets and damage to buildings and infrastructure) from extreme weather events are generally estimated as a matter of course, the value of indirect costs, for example due to travel disruptions, service interruptions or loss of business revenues, are less easily, and therefore less frequently, quantified. The costs of these indirect damages are the subject of a current research project funded by the EPA and based at UCC.(*) Here we describe some preliminary results from our analysis of disruptions to the road network in Co. Galway during the winter flooding of 2015/16.(**)

Under normal circumstances, commuting in Ireland involves substantial costs, in the form of the monetary costs of travel (ticket prices or the cost of fuel and other running costs for car drivers), as well as the welfare cost of the lost time spent commuting – generally considered an unpleasant but necessary activity. Previous research by one of us and a number of co-authors estimated these combined commuting costs as equivalent to about 30% of daily wages for the average commuter in the commuter belt around Dublin, about 26% for the average commuter in Co. Galway and 20% in Co. Cork (see Vega et al., 2016, as reported here and summarised here [PDF]). These costs reflect in part the heavy reliance on private car as mode of transport (76% in the Greater Dublin Area, 95% elsewhere in the country), as well as recent patterns of spatial development. The very high levels of car dependence, especially outside of Dublin, also highlights our economy’s vulnerability to disruptions to the road network.

Our research makes use of road closure data collected by Galway County Council in the aftermath of Storm Desmond (December 2015). The data we use details road status (open, closed, passable, one-lane only) and is time stamped, enabling us to observe which segments of the road network were affected on a particular day at a particular time. The data covers a 17 work day period of disruption (9th December 2015 to 5th January 2016). This data is combined with the Open Street Map road network, CSO POWSCAR 2011 and the SMILE model (Simulated Model of the Irish Local Economy) (O’Donoghue et al., 2012). Using this combined dataset, we simulate for every commuter in Co.Galway their commuting time and cost of commute (combined monetary and time costs) under the status quo and separately for each day (morning and evening) of the flooding period.

fig1-direct-costs-per-day-800x600
Fig.1 Average commuting costs in Co.Galway without flooding
fig2-length-roads-flooded-800x600
Fig.2 Areas affected by flooding
fig3-additional-time-minutes-800x600
Fig.3 Additional time spent commuting due to flooding (average per commuter per day)
Fig.4
Fig.4 Average commuting cost of the flooding (monetary + time costs) per commuter

Observing the commuting pattern before the flooding event, it is evident that those on the outer Galway city commuting belt already have high commuting costs [Fig.1]. Some of these areas also overlap with the areas worst affected by the floods [Fig.2]. Direct commuting costs for the worst affected areas ranged from €278-€680 per commuter over the 17 day period. In terms of commuting times, for some commuters the flooding involved an extra 30-60 minutes per day travel time [Fig.3]. When we include the time costs, that is monetary compensation for the extra time spent in your car, the total extra cost of the disruption represents up to 38% of the working wage over the period of disruption, for the average commuter in the worst affected areas [Fig.4].

On aggregate, we estimate the total cost of the disruption to commuting in Co. Galway during the flooding at €3.8million. This estimate assumes that every commuter in Galway actually travelled to work each day during the flooding. However, our estimates are conservative in that our model cannot account for delays on non-flooded routes due to additional volumes of traffic displaced from flooded roads, or for disruptions to commuters travelling between Galway and origins/destinations outside the county. We also do not count any costs imposed on commercial vehicles, disruptions to business activity, supply chains etc.

While Storm Desmond was considered a 1-in-100 year event, a near-real time attribution analysis found that such extreme rainfall was already up to 40% more likely due to the effects of human-induced climate change experienced to date, i.e. events like this are now a 1-in-72 year event (van Oldenborgh et al., 2015[PDF]). With further warming, these risks will likely multiply (IPCC, 2012, 2013 [both PDF]). In short, we can expect events like this to occur with substantially greater frequency in future. Understanding the full economic costs of these extreme events is an important first step in preparing for a future with increased weather risk.


Disclaimer: This post represents the views of the authors alone, and does not represent the views of the Environmental Protection Agency.

(*) This research is funded by the Irish Environmental Protection Agency, and involves collaboration with the Grantham Research Institute on Climate change and the Environment at the London School of Economics.

(**) The results reported are from a piece of research co-authored by Paul Kilgarriff and Tom McDermott (both UCC), Amaya Vega (NUIG), Cathal O’Donoghue (NUIG) and Karyn Morrissey (Exeter), presented [PDF] recently at Regional Studies Association (RSA) Student and Early Career Conference 2016, Newcastle, UK.

Author: Tom McDermott

I am a lecturer at the School of Economics, UCC, a Principal Investigator at the Environmental Research Institute, UCC and a Visiting Fellow at the Grantham Research Institute on Climate Change and the Environment at the London School of Economics. My research is at the intersection of environmental and development economics. I am co-editor with Sam Fankhauser of The Economics of Climate Resilient Development (Edward Elgar, 2016). I am also Principal Investigator for a project funded by the Environmental Protection Agency on the economic impacts of flooding in Ireland, and options for managing flood risk, including an emphasis on the role of insurance.