Original Article 207

Economic impacts of landslides and floods

on a road network
Mike G. Winter 1, 2, *, David Peeling3, Derek Palmer 4, James Peeling3

1 Winter Associates, Kirknewton, Midlothian, U.K. formerly Transport Research Laboratory (TRL), Edinburgh, U.K.
2 University of Portsmouth, Portsmouth, U.K.
3 Transport Research Laboratory (TRL), Wokingham, U.K.
4 Formerly Transport Research Laboratory (TRL), Wokingham, U.K., now Independent Transport Economist, U.K.

* Corresponding author: [email protected]

ABSTRACT
Even in the absence of serious injuries and fatalities, landslide and flood events can have significant socio-economic impacts. These
include the severance of access to and from relatively remote communities for services and markets for goods; employment,
health and educational opportunities; and social activities. The economic impacts can be classified as: direct economic impacts,
direct consequential economic impacts, and indirect consequential economic impacts. In addition, the vulnerability shadow cast
can be extensive, and its geographical extent can be determined by the transport network rather than the relatively small footprint
of the event itself. Using a number of debris flow events and a flood event in Scotland this paper places values on the economic
impacts of landslides and floods. It also demonstrates the widespread impact of the events by means of the vulnerability shadow
that is cast.

KEYWORDS
landslides; floods; hazard; risk; economic; social

Received: 17 May 2019

Accepted: 15 October 2019
Published online: 18 November 2019

Winter, M. G., Peeling, D., Palmer, D., Peeling, J. (2019): Economic impacts of landslides and floods on a road network. AUC
Geographica 54(2), 207–220
https://doi.org/10.14712/23361980.2019.18

© 2019 The Authors. This is an open-access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0).
208 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

1. Introduction activities, in the form of both management and miti-

gation and. a study was commissioned to assess and
A series of debris flows associated with monthly make recommendations on potential landslide reme-
average rainfall substantially in excess of the norm diation actions at the Rest and be Thankful site (Anon.
occurred in Scotland in August 2004. The resulting 2013a; Winter, Corby 2012).
landslides affected important parts of the trunk (stra- The rainfall-induced landslides occur in thin
tegic) road network, linking both cities and smaller, deposits of weathering products and other debris
remote communities. Notable events occurred (Fig- overlying rock that frequently dips out of the slopes
ure 1) at the A83 between Glen Kinglas and north which may be at angles of up to around 36°. Detailed
of Cairndow (9 August), the A9 north of Dunkeld descriptions of events are available (Winter et al.
(11 August), and the A85 at Glen Ogle (18 August). 2006), as are approaches to hazard and risk assess-
While there were no major injuries, 57 people had ment (Winter et al. 2005, 2009, 2013), and approach-
to be airlifted to safety at A85 Glen Ogle when they es to management and mitigation (Winter 2014a,
became trapped between two major debris flows. 2016). The links with rainfall are described by Winter
The A83 Rest and be Thankful site, while not affect- et al. (2010, 2019).
ed in August 2004, has been extremely active in recent In the absence of serious injuries and fatalities to
years with multiple debris flow events and road asso- those involved, the primary impacts of these events
ciated closures; in particular events in 2007 (Figure 2), were economic and social. Such impacts include the
2008, 2009, 2011, 2012, 2014, 2015, 2017 and 2018
had an adverse effect on the travelling public. As
a result the area has become the focus of not only
concern but also of extensive landslide risk reduction

Fig. 1 Map showing the strategic road network in Scotland. The

locations of the three main debris flow event groups that affected
the trunk road network in Scotland in August 2004 (1, 2, 3) are
shown (from Winter et al. 2009). The locations of the later A83, 2007 Fig. 2 View of the debris flows above and below the A83 on the
and 2014, landslides are coincident with the 2004 event at this scale approach to the Rest and be Thankful (location approximately that
and the A77, A76, A71 Bellfield Junction flood event from 2012 is of 1b in Figure 1). The head scar is at approximately 370 m Above
also shown. The locations for the events correspond to those shown Ordnance Datum (AOD), the A83 at 240 m AOD and the Old Military
in Figures 4 (1b), 9 (2), 10 (3) and 11 (4). Road (OMR) at 180 m AOD.
Economic impacts of landslides and floods on a road network 209

cost of delays and diversion on transport networks were summarized by Winter, Bromhead (2012), in
and the severance of access to and from relatively three categories, as follows:
remote communities for services and markets for – direct economic impacts,
goods; employment, health and educational opportu- – direct consequential economic impacts,
nities; and social activities. – indirect consequential economic impacts.
In 2004 for example, the A83, carrying up to 5,000 Direct economic impacts: The direct costs of clean-
vehicles per day (all vehicles two-way, 24 hour annu- up and repair/replacement of lost/damaged infra-
al average daily traffic, AADT) was closed for slightly structure in the broadest sense and the costs of search
in excess of a day, the A9 (carrying 13,500 vehicles and rescue. These should be relatively easy to obtain
per day) was closed for two days prior to reopening, or estimate for any given event, provided that records
initially with single lane working under convoy, and are still available.
the A85 (carrying 5,600 vehicles per day) was closed Direct consequential economic impacts: These gen-
for four days. The traffic flow figures are for the most erally relate to ‘disruption to infrastructure’ and are
highly trafficked month of the year (July or August). really about loss of utility. For example, the costs of
Minimum flows occur in either January or February closing a road (or implementing single-lane convoy
and are roughly half those of the maxima reflect- working with traffic lights) for a given period with
ing the importance of tourism and related seasonal a given diversion, are relatively simple to estimate
industries to Scotland’s economy. Substantial disrup- using well-established models. The costs of fatal/
tion was thus experienced by local and tourist traffic, non-fatal injuries and other incident accident costs
and goods vehicles. may also be included here and may be taken (on
This paper describes a study to assess the econom- a societal basis) directly from published figures. While
ic impacts of selected debris flow events in Scotland, these are set out for the costs of road traffic accidents,
based on the scheme set-out by Winter and Brom- or indeed rail accidents, there seems to be no partic-
head (2012). The impacts of floods can be assessed ular reason why they should be radically different to
using the same set of principles and metrics and the those related to a landslide or flood as all such inci-
impacts of one such event are also reported here. dents are likely to include the recovery of casualties
from vehicles. Indeed, for events in which large num-
bers of casualties may be expected to occur, data for
2. Economic impacts rail accidents may be more appropriate.
Indirect consequential economic impacts: Often
Due to the major contribution that tourism makes to landslide events affect access to remote rural areas
Scotland’s economy the impacts of road closures can with economies that are based upon transport-de-
be particularly serious during the summer months, pendent activities, and thus the vulnerability can be
during which period debris flows usually occur in July extensive and is determined by the transport network
and August. Nevertheless, the impacts of any debris rather than the event itself. If a given route is closed
flow event occurring during the winter months, for a long period then confidence in, and the ongoing
between October and November, and in January when viability of, local business may be affected, for exam-
debris flow usually occurs, should not be underesti- ple. Manufacturing and agriculture (e.g. forestry in
mated and events are arguably more frequent during western Scotland), are a concern as access to markets
the winter. Not surprisingly, the debris flow events is constrained, the costs of access are increased and
described created a high level of interest in the media business profits are affected and short-term to long-
in addition to being seen as a key issue by politicians term viability may be adversely affected; in Jamaica
at both the local and national level. Indeed, the effects a landslide on the B1 route in the Blue Mountains
of such small events which may, at most, directly effectively severed local coffee production from the
affect a few tens of metres of road cast a considerably most direct route to the international market for that
broader vulnerability shadow (Winter, Bromhead high value product (Figure 3). Perhaps of even more
2012), which defines the geographical extent of the concern are the impacts on tourist (and other service
impacts of landslides and floods. The qualitative eco- economy) businesses. It is important to understand
nomic impacts of such landslide (and flood) events how the reluctance of visitors to travel to and with-
include: in areas affected by landslides or floods is affected
– the loss of utility of parts of the road network, after an event that has received publicity and/or
– the need to make often extensive detours in order caused casualties and how a period of inaccessibili-
to reach a destination, ty (reduced or complete) affects the short- and long-
– the severance of access to and from relatively term travel patterns to an area for tourist services.
remote communities for services and markets for Such costs form a fundamental element of the overall
goods; employment, health and educational oppor- economic impact on society of such events. They are
tunities; and social activities. thus important to governments as they should affect
The economic impacts of a landslide event that the case for the assignation of budgets to landslide
closes a road, and its associated vulnerability shadow, risk mitigation and remediation activities. However,
210 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

these are also the most difficult costs to determine as the indirect consequential impacts of such event are,
they are generally widely dispersed both geograph- at best, rare.
ically and socially. Additionally, in an environment The vulnerability shadow cast can be extensive and
in which compensation might be anticipated, albeit its geographical extent can be determined by the trans-
often erroneously, those that have the best data, the port network, including closures and diversionary
businesses affected by such events, are also those that routes, rather than the relatively small footprint of the
anticipate such compensatory events. event itself (Winter, Bromhead 2012). In the particular
Typically analyses of the economic impacts of case of the event at the A83 Rest and be Thankful in
landslides and other natural disasters focus on the October 2007 of the order of around 400 m3 of material
direct and/or indirect impacts, with the latter being was deposited at road level with a footprint that closed
generally analogous to direct consequential as set- a few tens of metres of the road (Winter 2014); the vul-
out above (Schuster 1996; Highland 2006; Hearn et nerability shadow can be estimated to be of the order
al. 2008; Chang, Nojima 2001; Bono, Gutiérrez 2011; of 2,800 km2 (total area approximately 3,500 km2,
Bil et al. 2015; Klose et al. 2015). Attempts to evaluate 20% allowed for areas of sea) (Figure 4) which is, for

Fig. 3 Landslide on the B1 road at

Section in Portland Parish, Jamaica. This
event severed much of the local coffee
production industry from the ports
used to ship the product to market.
(The picture is a photo-collage and
some distortion is inevitable.)

Fig. 4 A relatively small debris flow

event (blue rectangle) closed the A83
at the Rest and be Thankful in October
2007; the vulnerability shadow that was
cast (bounded in red) was extensive
(Winter 2014a; 2014b). See Figure 1
(1b) for the event location in the wider
geographical context. (Image based on
OS Route Planner 2016 Map. © Crown
Copyright. All rights reserved Scottish
Government 100020540, 2019.)
Economic impacts of landslides and floods on a road network 211

the purpose of comparison, approximately two-and-a- The economic impact and the vulnerability shadow
half times the total land area of Hong Kong SAR. are concepts that apply equally to other discrete cli-
The area has a population density of approxi- mate driven events, such as floods, that may close parts
mately 13 people/km2 (www.argyll-bute.gov.uk) and of the road network. Both landslides and floods are
the event thus had the potential to have had an eco- generally thought to be likely to increase in frequency
nomic impact upon up to approximately 36,400 peo- as a result of climate change (Galbraith et al. 2005;
ple in Argyll, Bute, plus any transient (e.g. tourist) Anon. 2011; Winter et al. 2010; Winter, Shearer 2013).
population. Notwithstanding the above, it is clear that for some
It is instructive to make some simple comparisons events the hazard itself, and not the transport net-
with Hong Kong SAR, which has an average population work or, more pointedly, its density, that determines
density of around 6,500 people/km2 (www.gov.hk). the location, shape and extent (morphology) of the
This dictates a much greater transport network den- vulnerability shadow. It is therefore important to rec-
sity. Thus, and purely for the sake of comparison, in ognise that the morphology of the vulnerability shad-
order to have an economic impact on the same number ow caused by other types of event (e.g. glacial lake
of people the vulnerability shadow cast need only be outburst floods), may be determined by the nature of
approximately 5.6 km2 (2 km by 2.8 km, for example). the hazard itself.
It is not suggested that the economic impacts would An example in which the hazard itself determines
be similar for events with vulnerability shadows of the vulnerability shadow is that of the Seti River debris
these diverse sizes in Argyll, Bute and Hong Kong. flow in Nepal (Figure 5). On 5 May 2012 a major event
However, it is clear that the low density/dispersed caused significant erosion and deposition in the river
network in Argyll, Bute dictates a large vulnerability channel over an approximate 40 km length. The event
shadow while the much more dense/less dispersed was thought to have resulted from a failed landslide
network in Hong Kong means that the vulnerability dam. However, subsequent inspection of satellite
shadow will be much less extensive, with the possible imagery and aerial photography (Petley, Stark 2012;
exception of events that affect critical infrastructure Petley 2014), and more detailed site inspection and
corridors, as more alternative routes will exist and investigation (Dahal , Bhandary 2013) led to a rather
will be more proximal to the event (Winter 2014b). different conclusion; that the event was a debris flow
As part of this work (Winter et al. 2018), the initiated by part of a 22 Mm3 rock avalanche origi-
vulnerability shadow has been evaluated using nating on the slopes of Annapurna IV entering the
knowledge of the local transport networks and the upper stream channel at high speed. An estimated
socio-economic activity associated with the network 71 people lost their lives at Kharapani, some 20 km
that has been built up over a period of 30 years. This north of Pokhara. The vulnerability shadow was con-
includes a holistic evaluation of major nodes, origins strained by the dimensions of the hazard flow within
and destinations and includes both experience and the stream channel, extending beyond these bounds
knowledge gleaned from formal surveys (e.g. Winter only where infrastructure was damaged, including the
et al. 2013). footbridge at Kharapani.

Fig. 5 Residents of Kharapani located

on the platform in the middle distance
on the Seti River, Nepal, were among
fatalities from the 5 May 2012
debris flow event. The abutment of
the suspended footbridge is on the
platform.
212 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

3. Methodology If a road is closed, either fully or partially, as a result

of an event some or all of the users of that route will
3.1 Direct Economic Impacts have to take an alternative, diversionary route, which
Direct economic impacts should be the most straight- may be significantly longer than the primary route.
forward to determine. Indeed, this has generally Even if no diversion is necessary, then the capacity of
proved to be the case with relatively recent events the road will likely be reduced (e.g. through a lane clo-
that occurred within the currency of existing net- sure or the imposition of a speed limit) meaning that
work operation contracts. Thus, data relating to the queues form, particularly at peak times, slowing the
2007 and 2014 A83 Rest and be Thankful events were traffic flow. These effects can significantly increase
readily available from the company responsible for road users’ journey times. It is also pertinent to note
the operation of roads in this north-west part of Scot- that such queues if formed adjacent to areas of high
land at the times of enquiry, as was data for the 2012 landslide susceptibility can significantly increase the
flooding event. risks to the affected road users.
Data on direct economic impacts was generally The QUADRO (QUeues And Delays at ROadworks)
collected from the records of the companies respon- model provides a method for assessing the costs
sible for operating the road network. However, data imposed on road users while roadworks are being
from less recent events such as the landslide events carried out, considering:
of 2004 (Winter et al. 2005, 2006, 2009) proved more – Delays to road users: the change in road users’
difficult to obtain largely as both the operators and journey times, priced using the value of their time
auditors had changed since the events occurred; as (e.g. cost to their employer’s business of the time
Highland (2006) points out past data are generally spent travelling during the working day) based on
labour intensive to retrieve. the type of vehicle, its occupants and trip purpose.
The passage of time and the consequent lack – Fuel carbon emissions: the change in carbon emis-
of readily available records limited the resolution sions due to vehicle fuel consumption, based on
and reliability of the data that can be obtained for average figures per litre of fuel burnt and costed
the 2004 events. What data has been obtained was using estimated abatement costs (see STAG and
derived from high level reporting documents to Scot- WebTAG) (Anon. 2012a, 2012b).
tish Ministers and Senior Civil Servants and covers all – Accident costs: the change in the occurrence of
three of the event groups from August 2004 (A83, A9 accidents, in terms of the additional delay caused
and A85). This data has been interpreted and broken and the direct costs (e.g. property damage, police
down to the best of the ability of the original authors time and insurance administration).
and editors of the Scottish Road Network Landslide The program contains a model for allocating traffic
Study reports (Winter et al. 2005, 2009). to the diversion route if the site becomes overloaded,
representing both the road users that queue through
3.2 Direct Consequential Economic Impacts the site and those that take an alternate route in the
Direct consequential economic impacts are related case of a partial closure. The details of QUADRO,
to ‘disruption to infrastructure’ and concern loss of including all assumptions made in its calculations, are
utility or the costs imposed on road users. For exam- provided in the manual (Anon. 2015).
ple, the costs of closing a road (or implementing sin- In order to model a road closure in QUADRO,
gle-lane working with traffic lights) for a given period a diversionary route must be defined. The QDIV
with a given diversion, are relatively straightforward (QUADRO Diversion) tool was used to model the
to estimate using well-established models. The costs standard diversionary routes used by the road
of fatal/non-fatal injuries and incident damage only operator.
accidents are also included here and may be taken QDIV requires each diversionary route to be
(on a societal basis) directly from published figures defined in terms of a set of links (each defined as rural,
(Anon. 2013b). While these are set out for the costs urban, sub-urban or small town) that can be com-
of road traffic accidents (figures are also available for bined in series and parallel to build up a network. For
rail accidents) there seems to be no particular reason each event, a simplified diversionary network sche-
why they should be radically different to those relat- matic was developed and Google Maps was used to
ed to a landslide or flood as both are likely to include measure the length of each link (Winter et al. 2018).
the recovery of casualties from vehicles. Indeed, for Traffic data, represented as annual average daily traf-
events involving large numbers of casualties, it can be fic (AADT), were sourced using data from the relevant
argued that data relating to railway accidents may be Road Administrations.
more appropriate. In all of the cases presented here Where information was not available (e.g. lane
the accidents have been non-injury accidents (dam- and verge widths), the default values suggested in
age only) and the numbers of vehicles involved have the QUADRO manual were adopted. Classified traf-
been estimated from contemporaneous photographs fic counts (i.e. split into different vehicle types), and
taken by the first author and others. therefore the proportion of heavy vehicles, were only
Economic impacts of landslides and floods on a road network 213

Table 1 Site parameters for the direct consequential economic impact analysis.

Number vehicles Traffic (AADT) Junction length

Event HGVs (%) Closure type(s) Closure duration
damaged (vehicles/day)2 (km)
August 2004: A83 Glen
1 5,554 9 20 Full closure 2 days
Kinglas to Cairndow (L)1
August 2004: A9 N of Full closure, then shuttle 2 days full
5 13,864 18 18
Dunkeld (L) working with convoy 6 days convoy
August 2004: A85 Glen
3 4,403 10 26 Full closure 4 days
Ogle (L)
October 2007: A83 Rest Full closure, then shuttle 15 days full
1 5,748 10 20
and be Thankful (L) working3 27 days shuttle4
A77: Full closure 25 hours
September 2012: A77,
A76: Full closure then 19 hours full
A76, A71 Bellfield Junction 0 6,400 to 13,800 6 0.3 to 4.9
shuttle working 2 days shuttle
(F)
A71 Full closure 67 hours
Full closure (daytime) 10 hours
Full closure (night time) 28 hours
Convoy (daytime)6 42 hours
October 2014: A83 Rest
0 5,460 9.5 20 / 45 Convoy (night time)6 56 hours
and be Thankful (L)
Shuttle (daytime) 40 hours
Shuttle (night time) 42 hours
Shuttle (24 hours) 682 hours
1 L = Landslide; F = Flood.
2 Peak monthly figures, usually for August.
3 Single-lane working with traffic light control.
4 This figure represents the duration of the closure due to the instability and the immediate engineering works required to allow the reopening of the

road. It is acknowledged that the road was subsequently subject to single lane working with traffic light control for a significantly longer period due
to engineering works necessitated by the combination of this and subsequent events in the immediate vicinity.
5 20 km for full closure, 4 km for convoy working on the old Military Road (OMR).
6 Using the OMR, the temporary diversion used when the A83 Rest and be Thankful road is closed.

available for some links; either the proportion from Surveys of businesses were undertaken in the areas
the closest link or a nominal 10% HGVs was assumed of the 2004 A85 landslide; 2012 A77, A76, A71 Bell-
(Table 1). field Junction flood; and A83 2014 landslide events.
It was assumed that all of the roads affected were The surveys used questionnaires based on the Stat-
rural all-purpose single carriageways with a speed lim- ed Preference approach that were mailed to respon-
it of 96 km/h (60 mph), reduced to 48 km/h (30 mph) dents with follow up telephone calls to improve the
where part of the road remained open following the response rate that varied between 20.8% (A5 Glen
landslide or flood event, and that the length of the Ogle) 17% (A83 Rest and be Thankful) and 11.7%
affected site in each case was 100 m. (The speed limit (A77, A76, A71 Bellfield Junction).
at the A85 site was reduced from the national speed
limit of 96 km/h (60 mph) to 80 km/h (50 mph) in
November 2015, more than a decade after the events 4. Results
and around the time that aspect of the study was
completed.) 4.1 Direct Economic Impacts
The available data is reported in Table 2, adjusted to
3.3 Indirect Consequential Economic Impacts 2012 prices.
There is a wide range of possible approaches to esti- Direct economic impacts include:
mating the indirect consequential economic impacts 1. The direct costs of clean-up and the costs of search
of landslides. These include: and rescue.
– cost-benefit analysis, 2. The repair/replacement of lost/damaged infra-
– cost-effectiveness analysis, structure in the broadest sense.
– willingness to pay, These might otherwise be described as ‘emergency
– multi-criteria analysis, response’ and ‘remedial works’, respectively as in Table 2.
– methods based upon Transport Appraisal. Direct economic costs for the landside events
In addition, there are bespoke methods designed range between approximately £250k and £1,700k.
to address a particular set of circumstances (McLeod For the flood event in a more developed peri-urban
et al. 2005; Anon. 2013b) as described by Winter et part of Scotland the direct costs were relatively small
al. (2018). (around £25k).
214 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

Table 2 Direct economic impacts (at 2012 prices).

Event Emergency response Remedial works Total

August 2004: A83 Glen Kinglas to Cairndow £395,043 £395,043
August 2004: A9 N of Dunkeld £921,766 £921,766
August 2004: A85 Glen Ogle £658,405 £658,405
October 2007: A83 Rest and be Thankful £320,772 £1,372,6291 £1,693,401
September 2012: A77, A76, A71 Bellfield Junction £16,756 £8,333 £25,088
October 2014: A83 Rest and be Thankful £245,328 £02 £245,328
1  omprises: debris barrier (including design, tender, award, construction and supervision) £425,446 (at 2012 prices); works prior to culvert
C
replacement (including ongoing traffic management and carriageway protection, culvert design, geotechnical design and certification, ground
investigation and diversion fibre optic telecommunications cable) £181,184; and culvert replacement (construction) £765,999. These works were
undertaken in direct response to the October 2007 event and may thus be attributed to this event.
2 While an extensive programme of remedial works has been undertaken at the wider A83 Rest and be Thankful site, some of which is focused on the

location of the 28 October 2014 event, this was installed prior to the event and thus cannot be attributed to this event. In broad terms the remedial
measures, including both management and mitigation measures (Winter 2014), worked as anticipated and would not have been expected to prevent
a debris flow of this size from reaching the road.

4.2 Direct Consequential Economic Impacts (Anon. 2015) and the associated values are based on
QUADRO enables the calculation of the costs of user national statistics.
delays and diversions, carbon emissions from vehi- Careful consideration of the relative traffic lev-
cles and accidents associated with the road works, els, and closure type and duration (Table 4), reveals
reporting the costs on the basis of an average day patterns that are broadly consistent with those that
over a whole week (Table 3); the total costs for each might be inferred intuitively, as follows:
site, taking into account the duration of the impacts, – The costs of similar closures depend on traffic levels;
are summarised in Table 4. Implicit are assumptions costs being in proportion to traffic (A9 cf. A83 2004).
regarding the costs of time (vehicle occupancy, jour- – Doubling the duration incurs higher costs, but may
ney purpose, and the value of time for both occupants be reduced if the traffic levels are lower (A83 2004
and vehicles), vehicle operating costs (and associated cf. A85).
carbon costs), and the value of accidents that occur – A much longer duration increases the costs signifi-
within the section(s) of road under consideration cantly (A83 2007).

Table 3 Incident accident costs (per vehicle) and QUADRO daily closure costs (at 2012 prices).

October 2007: September 2012:

August 2004: October 2014: A83
August 2004: A83 A83 Rest and be A77, A76, A71
A9 N of Dunkeld August 2004: Rest and be Thankful1
Cost (£) Glen Kinglas to Thankful Bellfield Junction
(Full closure/ A85 Glen Ogle (Full closure/ convoy/
Cairndow (Full closure/ (Full closure/
shuttle working) shuttle working)
shuttle working) shuttle working)
Accident incident cost 2,520 2,520 2,520 2,520 2,520 2,250
Delay cost (daily) 84,071 270,885 / 135,339 71,679 88,040 / 461 1,548,624 / 94,363 168 to 83,427
Carbon cost (daily) 6,380 18,608 / 9,304 6,629 6,590 / 6 73,922 / 671 2 to 6,262
Accident cost (daily) −4,360 −11,254 / −5,627 −4,494 −4,512 / 794 −38,568 / 7,564 −4,268 to 312
1 T he daily delay, carbon and accident costs for the October 2014 Rest and be Thankful event vary significantly, by several orders of magnitude, for the
various types of closure described in Table 2 and accordingly only the range is reported here. Full details are given by Winter et al. (2018). The higher
costs (lower daily accident costs) are for full closure and the lowest costs (highest daily accident costs) are for shuttle working.

Table 4 Total incident accident costs and QUADRO total closure costs (at 2012 prices).

October 2007: September 2012: October 2014: A83

August 2004:
August 2004: A83 A83 Rest and be A77, A76, A71 Rest and be Thankful
A9 N of Dunkeld August 2004:
Cost (£) Glen Kinglas to Thankful Bellfield Junction (Full closure/
(Full closure/ A85 Glen Ogle
Cairndow (Full closure/ (Full closure/ convoy/shuttle
shuttle working)
shuttle working) shuttle working) working)
Accident incident cost 2,520 12,600 7,560 2,520 0 0
Delay cost 168,143 1,218,460 286,718 1,333,020 3,080,542 173,956
Carbon cost 12,762 83,737 26,514 99,029 151,669 9,164
Accident cost −8,721 −45,288 −17,974 −46,247 −71,309 29,466
Total 174,703 1,269,508 302,817 1,388,322 3,160,902 212,587
Economic impacts of landslides and floods on a road network 215

Fig. 6 Word map of responses from survey respondents: A85 Fig. 7 Word map of responses from survey respondents: A77, A76,
Glen Ogle, 18 August 2004 landslide. A71 Bellfield Junction Flooding: 21 September 2012.

Of particular interest are the negative costs (i.e. relate to the effects, risks, or impacts, that derive from
cost reductions) for traffic accidents during post- the event. In this case the most frequently used word
event diversions and/or restricted traffic flow. These is ‘road’, with words such as ‘closed’, ‘staff’, ‘visitors’,
reduced accident costs suggest a decrease in accident ‘due’, ‘access’, ‘tourism’, ‘minor’ and ‘island’ also coming
numbers and/or accident severity and seem most to the fore. These latter responses seemingly describe
likely to be as a result of reduced traffic speeds lead- the consequences of the hazard, or the economic risks
ing to an increased opportunity to avoid accidents and associated with the hazard, rather than the hazard
lower severity when they do occur. itself, implying a greater economic impact or, at least,
The landslide events were located in rural areas a greater awareness of the economic impact.
and their impacts are upon those areas and small
towns and villages. Direct consequential costs range
between approximately £180k and £1,400k for the
landslide events. The latter costs are largely depend-
ent upon the amount of traffic that uses the road and
the duration of the disruption.
The flood event was located in a much more devel-
oped part of Scotland and on the edge of a town (Kil-
marnock). The peri-urban flood location and much
shorter event duration, places a different complex-
ion on the direct consequential economic impacts
which were more than twice those for the A83 2007
(c. £3,200k). Fig. 8 Word map of responses from survey respondents: A83 Rest
and be Thankful, 28 October 2014 landslide.
Notwithstanding this the impacts of the A83
event(s) should not be underestimated: those impacts
were borne by a much smaller number of people over 5. Vulnerability shadow
an extended period; the impacts on individuals and
individual businesses seem likely to have been con- The vulnerability shadow for the October 2007 debris
siderably greater. This part of the analysis also does flow is described in Section 2 (Figure 4) and is esti-
not take account of the longer term indirect conse- mated at around 2,800 km2. This description holds
quential economic impacts (see Section 4.3). for the event in 2014 also as the road closure was on
the same link between the junctions with the A814
4.3 Indirect Consequential Economic Impacts at Arrochar and the B828 immediately to the north
The surveys of businesses in the areas of these events of the Rest and be Thankful car park with no entry or
provided cost information that could be interpreted exit routes to the A83 between.
in a number of ways and therefore gave a very wide While the 2007 and 2014 A83 debris flow events
range of potential results. The results did, however, occurred to the east of the B828 (serving Lochgoil-
provide useful qualitative information (Winter et al. head), the August 2004 events at the A83 were locat-
2018). For events of lesser impact, descriptors that ed to the west of the B828 in Glen Kinglas and fur-
relate to the hazard are used: ‘landslide’, ‘flooding’ and ther to the west, beyond the A815 (serving inter alia
other words that describe the event itself are also to Dunoon) to the west of Cairndow (around 5 km to the
the fore (Figures 6 and 7). north-west of the 2007/2014 events shown in Figure
In contrast responses to events of greater impact 4). Notwithstanding this, the differences in the diver-
and or repetition such as at the A83 (Figure 8), at sions for the two sets of events were subtle and it is
which a significant number of events and consequent broadly considered that the extent of the vulnerability
closures have occurred over the past 20 years, tend to shadow was not dissimilar.
216 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

Fig. 9 Three main debris flow events (blue

rectangle) closed the A9 N of Dunkeld
on 11 August 2004; the vulnerability
shadow that was cast (bounded in red)
was extensive and reflects the importance
of the A9, particularly to the communities
to the north. See Figure 1 (2) for the
event location in the wider geographical
context. (Image based on OS Route Planner
2016 Map. © Crown Copyright. All rights
reserved Scottish Government 100020540,
2019.)

Fig. 10 Two debris flow events (blue rectangle) closed the A85 in Glen Ogle on 18 August 2004; the vulnerability shadow that was cast
(bounded in red) was limited by the reasonably good range of alternative routes in the area. See Figure 1 (3) for the event location in the
wider geographical context. (Image based on OS Route Planner 2016 Map. © Crown Copyright. All rights reserved Scottish Government
100020540, 2019.)
Economic impacts of landslides and floods on a road network 217

Fig. 11 Flooding at the A77, A76, A71 Bellfield Junction (blue rectangle) closed all three roads on 21 September 2012; the area of the
vulnerability shadow that was cast (bounded in red) was limited by the availability of diversionary routes. See Figure 1 (4) for the event
location in the wider geographical context. (Image based on OS 1:250,000 mapping. © Crown Copyright. All rights reserved Scottish
Government 100020540, 2019.)

Those for the other 2004 events were estimat- records of landslides and their associated costs were
ed at around 8,000 km2 (A9: Figure 9) and 800 km2 inadequate and that a standardised approach to
(A85: Figure 10). The vulnerability shadow cast for landslide loss inventory would reduce the cost, and
the A9 debris flow events to the north of Dunkeld improve the usability and availability of landslide loss
(11 August 2004) is significant (Figure 9), reflecting information. Further work by Highland (2012) deals
the importance of the A9 as the primary north-south in more detail with both direct economic impacts and
route in Scotland, leading to the most northerly city also some aspects of direct consequential and indi-
of Inverness (population around 47,000) and numer- rect consequential economic impacts, although these
ous smaller communities along the route and to the latter two are usually described in a qualitative man-
north of Inverness, and the relative paucity of alterna- ner. Highland describes decreased economic activi-
tive routes and their commensurate length. The A85 ty in some areas and increased economic activity in
Glen Ogle vulnerability shadow cast for the 18 August other areas (e.g. as a result of the Glenwood Canyon
2004 debris flow is, by comparison, relatively small Rockfall) and both higher than usual flight costs and
(Figure 10). This, in turn, reflects the relatively cen- an increase in the incidence of charter aeroplane use
tral position of that area in Scotland and the existence between key locations on either side of the event.
and relative efficiency of alternate routes through and A related study (Klose et al. 2015) collected data
around the area. and modelled the costs incurred due to landslides
The vulnerability shadow cast by the 21 Septem- that affected the federal road network in the Lower
ber 2012 flooding event at the A77, A76, A71 Bell- Saxon Uplands in NW Germany. The approach used
field Junction (Figure 11) was smaller still at around the data collected at a local level to extrapolate direct
500 km2 (the nearby town of Kilmarnock has a popu- costs for the region, using the results of a susceptibil-
lation of approximately 46,000). The road network in ity assessment and an infrastructure exposure model.
this area is much denser, and the population density is However, this study deals only with direct economic
much higher, than that in the other areas studied and impacts and this degree of extrapolation across the
this leads to reasonable alternative routes, albeit with network was not considered appropriate to the sit-
significant associated disruption to traffic. uation in Scotland where landslides occur relatively
infrequently.
Hearn et al. (2008) deal with both direct and direct
6. Discussion consequential economic impacts, but their approach
to direct consequential impacts appears to be based
Highland (2006) (see also Schuster 1996; Schus- on the assumption that all vehicles (and drivers and
ter, Highland 2007; Highland 2012) concluded that passengers) that would normally use the road for the
218 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

period of the closure will wait for it to be reopened. of a route study of the A83 (Anon. 2013a) returning
While this might be appropriate for Laos, where the a figure of £286k per landslide at the A83 Rest and
study was conducted, and reflects the morphology of be Thankful. While this provides useful context, it is
the road network it does not account for either diver- not entirely clear how much of that total would be
sions or restrictions of traffic flow as are more normal accounted for by direct consequential impacts that
in Europe. were also included or for how long the impacts were
The lack of a robust database of landslide events assumed to endure.
makes a quantitative national economic analysis The vulnerability shadow proved to be a useful
difficult. Almost 40 landslides are reported in the tool to understand and articulate the extent of the
media in the UK annually (38 in 2010) (Gibson et al. socio-economic vulnerabilities and, indeed, the pop-
2012). Anon. (2013a) noted that between 1 January ulations potentially affected.
2007 and 31 October 2012, the A83 at the Rest and The vulnerability shadows determined from the
Be Thankful was closed five times as a result of land- various events demonstrate that their extent is deter-
slides and these and other landslides have been enu- mined by the density of the network, and the linked
merated and described (e.g. Winter et al. 2005, 2006, availability and suitability of alternative routes. Clear-
2009) extensively in recent years. Despite this Dobbie ly there is also a strong correlation between these
et al. (2011) consider that landslides are rare in Scot- factors and the population served by the routes and
land. Whatever the rarity or otherwise of such events, affected by the events; as the density of the network,
better data is critical to the effective management of and availability and suitability of alternative routes
risks, as highlighted for flood hazards and risks by the reduces so does the population served. This must of
Association of British Insurers (Anon. 2010). course be set against the fact that the lower density
Importantly, this current work confirms High- networks are often lifeline routes with few, if any,
land’s (2006) assertion that past data for direct alternatives and their importance to a smaller popu-
economic impacts are generally labour intensive to lation is amplified. The authors are not aware of this
retrieve. The experience here has been that as people or a similar tool being used in other studies.
move on both knowledge and experience are lost but,
even more critically, as contracts pass to new organi-
zations data and information about events is lost. 7. Summary and conclusions
The modelling of direct consequential econom-
ic impacts has not been attempted before and this This paper presents the results of a study to develop
throws up some interesting but rather unsurprising methods of obtaining data on the economic impacts of
conclusions – not least that the total costs depend landslides and the associated extent of those impacts.
primarily on the amount of traffic affected and the The economic impacts of landslides are considered
duration for which it is affected. Notwithstanding this, in three categories: direct economic impacts, direct
it is important to recognize that where traffic levels consequential economic impacts, and indirect con-
are lower the network may represent a lifeline route sequential economic impacts. This approach is also
with limited, complex/lengthy and/or no alternative applicable to events that reflect relatively discrete
routes. closures including climate-driven flooding.
Similarly, indirect consequential economic impacts The work presented herein includes data for five
are rarely evaluated. The surveys conducted for this Scottish landslide events that occurred between 2004
work provided important and very useful qualitative and 2014. Direct costs range between approximate-
data as set-out in Section 4.3 and represented in Fig- ly £250k and £1,700k for the landslide event, while
ures 6,7 and 8. Indeed, some of the comments from the those costs for the flood event are relatively small.
A85 Glen Ogle survey confirmed Highland’s (2006) Direct consequential costs range between around
assertion that economic impacts need not only be £180k and £1,400k for the landslide events. The lat-
negative, that some can be positive: ter are largely dependent upon the amount of traffic
– “More people rented bikes to go and visit the land- that uses the road and the duration of the disruption.
slide site. The landslide itself had little impact, it For a flood event in a more developed peri-urban
was the associated bad weather that was the prob- part of Scotland although the direct costs were small
lem - the main street was flooded” according to one but the direct consequential costs (c. £3,200k) much
retailer. greater than for any of the landslide sites considered.
– Hoteliers also seem to have benefited as “[they were It is also worth noting that flood event was of a rela-
able to] put up people in the guest house who had tively short duration compared to the high cost land-
been trapped in the glen (and who had been air- slide event.
lifted out) free of charge. The same happened in Work on indirect consequential impacts has pro-
different places around Killin,” which apparently vided valuable qualitative insights although meaning-
“improved the image of the town.” ful quantitative data proved rather elusive.
Notwithstanding this an assessment that included The vulnerability shadow proved to be a useful
indirect consequential impacts was reported as part tool to understand and articulate the extent of the
Economic impacts of landslides and floods on a road network 219

socio-economic vulnerabilities and the populations Government, Edinburgh. (Accessed December 2012:
potentially affected. They also aid in understanding www.sepa.org.uk/land/land_publications.aspx.)
and communicating the inter-relations between the Galbraith, R. M., Price, D. J., Shackman, L. (Eds.) (2005):
density of the network, the linked availability and Scottish road network climate change study. Scottish
Executive, Edinburgh.
suitability of alternative routes.
Gibson, A. D., Culshaw, M. G., Dashwood, C., Pennington,
C. V. L. (2012): Landslide management in the UK
the problem of managing hazards in a ‘low-risk’
Acknowledgements environment. Landslides 10(5), 599–610, https://
doi.org/10.1007/s10346-012-0346-4.
Transport Scotland’s funding of this work is gratefully Hearn, G., Hunt, T., Aubert, J., Howell, J. (2008): Landslide
acknowledged. impacts on the road network of Lao PDR and the
This paper is published with the permission of feasibility of implementing a slope management
Transport Scotland and TRL Limited. programme. Proceedings, International Conference on
Management of Landslide Hazard in the Asia-Pacific
Region (Ed: Chigira, M.), 187-195. The Japan Landslide
References Society, Tokyo.
Highland, L. M. (2006): Estimating landslide
Anon. (2010): Fighting flood risk together. Association of losses – preliminary results of a seven-state pilot
British Insurers, London. project. US Geological Survey Open File Report
Anon. (2011): Scottish road network climate change 2006–1032, USGS, Reston, VA, https://doi.org/10.3133
study: UKCP09 update. Transport Scotland, Edinburgh. /ofr20061032.
(Accessed February 2014: http://www Highland, L. M. (2012): Landslides in Colorado, USA:
.transportscotland.gov.uk/.) impacts and loss estimation for the year 2010. US
Anon. (2012a): Scottish Transport Appraisal Guidance. Geological Survey Open File Report 2012–1204, USGS,
Transport Scotland, Edinburgh. (Accessed February Reston, VA, https://doi.org/10.3133
2014: http://www.transportscotland.gov.uk/stag /ofr20121204.
/home.) Klose, M., Damn, B., Terhorst, B. (2015): Landslide
Anon. (2012b): TAG UNIT 3.3.5: The Greenhouse Gases Sub- cost modelling for transportation infrastructures:
Objective, August. Department for Transport, London. a methodological approach, Landslides, 12, 321–334,
(Accessed February 2014: http://www.dft.gov.uk https://doi.org/10.1007/s10346-014-0481-1.
/webtag.) MacLeod, A., Hofmeister, R. J. Wang, Y., Burns, S. (2005):
Anon. (2013a): A83 Trunk Road Route Study: Part A – A83 Landslide indirect losses: methods and case studies
Rest and be Thankful. Final Report. Report prepared by from Oregon, State of Oregon, Department of Geology
Jacobs for Transport Scotland, 212p. (Accessed February and Mineral Industries Open File Report O-05-X,
2014: http://www.transportscotland.gov.uk/road Portland, OR.
/maintenance/landslides.) Petley, D. N. (2014): The Seti River debris flow in Nepal –
Anon. (2013b): Reported road casualties Scotland 2012. what was the role of the smaller landslide downstream?
Transport Scotland, Edinburgh. (Accessed September (Accessed February 2014: blogs.agu.org/landslideblog
2015: www.transportscotland.gov.uk/statistics /2014/02/07/seti-river/.)
/reported-road-casualties-scotland-all-editions.) Petley, D. N., Stark, C. (2012): Understanding the Seti
Anon. (2015): The QUADRO 4 Manual, Part 2: The valuation River landslide in Nepal. (Accessed February 2014:
of costs in QUADRO. (Accessed September 2015: http:// blogs.agu.org/landslideblog/2012/05/23
tamesoftware.co.uk/manuals/manuals.html/.) /understanding-the-seti-river-landslide-in-nepal/.)
Bono F., Gutiérrez E. (2011): A network-based analysis of Schuster, R. L. (1996): Socioeconomic significance of
the impact of structural damage on urban accessibility landslides, Landslides – Investigation and Mitigation:
following a disaster: the case of the seismically damaged Transportation Research Board Special Report 247,
Port Au Prince and Carrefour urban road networks. 36–75, Washington, DC.
Journal of Transport Geography 19, 1443–1455, https:// Schuster, R. L., Highland, L. M. (2007): The Third Hans Cloos
doi.org/10.1016/j.jtrangeo.2011.08.002. Lecture, Urban landslides: socioeconomic impacts and
Bíl, M., Vodák, R., Kubeček, J., Bílová, M., Sedoník, J. (2015): overview of mitigative strategies. Bulletin of Engineering
Evaluating road network damage caused by natural Geology and the Environment 66, 1–27, https://
disasters in the Czech Republic between 1997 and 2010. doi.org/10.1007/s10064-006-0080-z.
Transportation Research Part A: Policy and Practice 80, Winter, M. G. (2014a): A strategic approach to landslide
90–103, https://doi.org/10.1016/j.tra.2015.07.006. risk reduction. International Journal of Landslide and
Chang, S.E., Nojima, N. (2001): Measuring post-disaster Environment 2, 14–23.
transportation system performance: the 1995 Kobe Winter, M. G. (2014b): The vulnerability shadow cast by
earthquake in comparative perspective. Transportation debris flow events, Engineering Geology for Society
Research Part A: Policy and Practice 35, 475–494, and Territory, Volume 6: Applied Geology for Major
https://doi.org/10.1016/S0965-8564(00)00003-3. Engineering Works (Eds: Lollino, G., Giordan, D., Thuro,
Dahal, R. K., Bhandary, N. P. (2013): Excursion guidebook L., Carranza-Torres, C., Wu, F., Marinos, P. Delgado, C.),
for Pokhara Valley area. Unpublished. 641–644, Heidelberg: Springer.
Dobbie, K. E., Bruneau, P. M. C., Towers, W. (Editors) (2011): Winter, M. G. (2016): A strategic approach to debris
The state of Scotland’s soil. Natural Scotland, Scottish flow risk reduction on the road network. Procedia
220 Mike G. Winter, David Peeling, Derek Palmer, James Peeling

Engineering 143, 759–768, https://doi.org/10.1016 Winter, M. G., Macgregor, F., Shackman L. (Eds.) (2009):
/j.proeng.2016.06.121. Scottish road network landslides study: implementation.
Winter, M. G., Bromhead, E. N. (2012): Landslide risk: Transport Scotland, Edinburgh.
some issues that determine societal acceptance. Natural Winter, M. G., Dent, J., Macgregor, F., Dempsey, P., Motion, A.,
Hazards 62, 169–187, https://doi.org/10.1007 Shackman, L. (2010): Debris flow, rainfall and climate
/s11069-011-9987-1. change in Scotland. Quarterly Journal of Engineering
Winter, M. G., Corby, A. (2012): A83 Rest and be Thankful: Geology and Hydrogeology 43, 429–446, https://doi
ecological and related landslide mitigation options. .org/10.1144/1470-9236/08-108.
Published Project Report PPR 636, Transport Research Winter, M. G., Harrison, M., Macgregor, F., Shackman, L.
Laboratory, Wokingham. (2013): Landslide hazard assessment and ranking on
Winter, M. G., Shearer, B. (2013): Climate change and the Scottish road network. Proceedings, Institution of
landslide hazard and risk – a Scottish perspective. Civil Engineers (Geotechnical Engineering) 166(GE6),
Published Project Report PPR 650, Transport Research 522–539, https://doi.org/10.1680/geng.12.00063.
Laboratory, Wokingham. Winter, M. G., Shearer, B., Palmer, D., Peeling, D., Peeling,
Winter, M. G., Macgregor, F., Shackman, L. (Eds.) (2005): J., Harmer, C., Sharpe, J. (2018): Assessment of the
Scottish Road Network Landslides Study. The Scottish economic impacts of landslides and other climate-driven
Executive, Edinburgh. events. Published Project Report PPR, TRL,
Winter, M. G., Heald, A., Parsons, J., Shackman, Wokingham.
L., Macgregor, F. (2006): Scottish debris flow events Winter, M. G., Ognissanto, F., Martin, L. A. (2019): Rainfall
of August 2004. Quarterly Journal of Engineering thresholds for landslides: deterministic and probabilistic
Geology and Hydrogeology 39, 73–78, https://doi approaches. Published Project Report PPR 901,
.org/10.1144/1470-9236/05-049. Transport Research Laboratory, Wokingham.