Pregnolato, M., Winter, A. O., Sen, A. D., Mascarenas, D., Bates, P.

D., & Motley, M. R. (2021). An integrated impact analysis for riverine

bridges subjected to high river flows. In H. Yokota, & D. M. Frangopol
(Eds.), Bridge Maintenance, Safety, Management, Life-Cycle
Sustainability and Innovations: Proceedings of the Tenth International
Conference on Bridge Maintenance, Safety and Management
(IABMAS 2020), Sapporo, Japan, 11-15 April 2021 (1 ed., pp. 1258-
1264). CRC Press/Balkema, Taylor & Francis Group.
https://doi.org/10.1201/9780429279119
Peer reviewed version

Link to published version (if available):

10.1201/9780429279119

Link to publication record on the Bristol Research Portal

PDF-document

This is the author accepted manuscript (AAM). The final published version (version of record) is available online
via Taylor & Francis at https://doi.org/10.1201/9780429279119 . Please refer to any applicable terms of use of
the publisher.

University of Bristol – Bristol Research Portal

General rights

This document is made available in accordance with publisher policies. Please cite only the
published version using the reference above. Full terms of use are available:
http://www.bristol.ac.uk/red/research-policy/pure/user-guides/brp-terms/
An integrated impact analysis for riverine bridges subjected to high river
flows
M. Pregnolato, P. Bates
University of Bristol, Bristol, United Kingdom
Andrew O. Winter, Dakota Mascarenas, Andrew D. Sen, Michael R. Motley
Dep. of Civil and Environmental Engineering, University of Washington, Seattle, USA

ABSTRACT: Flood events are the most frequent cause of damage to infrastructure compared to any other
natural hazard, and global changes (climate, socio-economic, technological) are likely to increase this damage.
Transportation infrastructure systems are responsible for moving people, goods and services, and ensuring con-
nection within and among urban areas; a failed link in this system can impact the community by reducing
evacuation capability, recovery operations and the overall economy. Bridges are critical links in the system,
since they are associated with less redundancy and a high construction cost. Riverine bridges are particularly
prone to failure during flood events; in fact, the risks to bridges from high river flows and bank erosion have
been recognized as crucial at global level. This study aims to establish rigorous practices of computational fluid
dynamics (CFD) for modelling hydrodynamic forces on inundated bridges, and understanding the consequences
of such impact on the surrounding network. Objectives of this study are to model hydrodynamic forces as
demand on the bridge structure and to advance a reliability analysis of the structure under the modelled loading.
Implications of the hydrodynamic impact on the performance and functionality of the surrounding transport
network are discussed. This research will help to fill the gap between current guidance for design and assess-
ment of bridges relevance within the overall transport system.

1 INTRODUCTION Riverine bridges are intrinsically vulnerable to

flooding, as located in the area of the riverbed. Scour
Bridges are crucial elements of the transport net- is the most common cause of bridge failure, however
work given their high construction costs and the lack hydrodynamic failure could be as critical for bridge
of alternatives routes. They underpin the surrounding piers on bedrock (where scour is unlikely), and for the
roads, thus the service disruption caused by a bridge decks of all flooded bridges (Oudenbroek et al.,
can knock-out communities’ access and connections, 2018). In terms of consequences, natural hazards can
impair emergency planning and evacuation routes, damage bridges structurally (thus causing direct
impact economies and businesses. Bridges can be dis- physical damages), but also cause travel time delays
rupted by a series of man-made and natural events; in and rerouting that lead to indirect losses.
particular, flood and scour represent the most fre-
quent cause of bridge failures (>50% of all failures; Scour damage is out of the scope of this paper and
Wardhana and Hadipriono, 2003). wide literature is already available (e.g. Ko et al.,
2014; AASHTO, 2002). On the contrary, literature
Flooding is a natural phenomenon whose impacts about bridges impacted by hydrodynamic forces due
have been exacerbated in recent years by urbanisation to riverine floods is limited, especially concerning
(e.g. increase of impermeable surfaces), inappropriate fragility models or reliability analysis (Gidaris et al.,
land use in flood-prone areas and climate change. 2016).
Rainfall events are becoming more frequent and in- There is some research dedicated to tsunami impact
tense (Solomon et al., 2007), triggering bridge inci- to bridges (Motley and al., 2015; Lomonaco et al.
dents and failures all over the world (Cumbria, UK, 2018; Qin et al., 2016; Winter et al., 2017), where
2009; Drake, Colorado, 2013; Texas, 2018). For ex- computational fluid dynamics (CFD) techniques are
ample, most recently, Grinton Bridge in Yorkshire used to compute hydrodynamic forces on bridges and
(North-West UK) was washed away by floodwaters components. Also, Kerenyi et al. (2009) applying
after 82.2 mm of rain fell in 24 hours on the 31st of CFD to compute hydrodynamic forces on inundated
July 2019. bridge decks, however the analysis is limited to the
evaluation of drag and lift forces, without investigat- (https://www.openfoam.com/). This second steps
ing impact and consequences. Multi-hazard studies model flood flows for considered flooding scenario
have investigated the interaction and implication of (return period).
multiple hazards acting on a single structure (Gidaris
et al., 2016; Carey et al., 2019; Trevor et al., 2019),
especially between earthquake and tsunami. Other
studies (Mondoro and Frangopol, 2018; Yilmaz et al.,
2015) that tackle flood impact to bridges generally
express the hazard through flood hazard curves, gen-
erated via flood-frequency analysis, while a detailed
hydraulic analysis is beyond the scope of the work.
No study is looking in a complete way to the prob-
lem of bridges impacted by high-river flows, account-
ing for the complexity of the hydrodynamic forces to
which the bridge is subjected. Moreover, the impact
of the reduced service on a bridge on the surrounding
network is almost inexistent in literature. Given this
limited availability of models, this paper aims to de- Figure 1. The methodological flowchart.
velop an integrated framework to assess the flooding
impact on riverine bridges from the structural- to the The third step is to determine the response of the
network-level. This research tackles several different bridge subjected to flood. Being a relatively low-
bridge configurations for varying condition (velocity complex but powerful tool, the OpenSees software
and depth) of flows as to understand the structural re- framework (McKenna et al. 2010) is adopted in this
sponse in a range of flooding conditions. This work is study. Mondoro and Frangopol (2018) describe sali-
novel since it represents a first tentative to couple ent limit states for bridges subjected to hydraulic
CFD with both finite element (FE) and network anal- loads, but the subset studied in this paper are shown
ysis for bridges subjected to flooding. Eventually, this in Fig. 2.
approach will be functional for understanding struc-
tural damage and functional loss for a range of
bridges, to assess risk and potential retrofitting
measures.

2 METHOD

This paper proposes a framework to assess the im-

Figure 2. Bridge limit states investigated due to flood load-
pact of high river flows to bridges and surrounding
ing.
roads (Fig. 1). The framework integrates into a com-
prehensive method that considers bridge susceptibil-
ity to hazard, reliability analysis and network-level A bridge could sustain (i) “catastrophic failure”,
impact. when the bridge deck, piers or foundation structurally
The first step is to determine the intensity measures fail; or (ii) “temporary failure”, when the floodwater
of the river flow, i.e. flow depth and velocity. For the reaches the deck and the bridge. Hydraulic pressures
simulation of fluvial flooding, the model (drag, lift and overturning moment) are assessed for
LISFLOOD-LP is used in this study. Bridge infor- potentially dislodging the deck from piers, when sub-
mation, such as geometry and design, can be retrieved merged or partially submerged, and overtopping of
through publicly available databases (if any, e.g. the the deck is evaluated qualitatively from the CFD
US National Bridge Inventory) or by dealing with lo- model. Though these limit states have significantly
cal infrastructure managers and authorities. In partic- different long-term consequences, both result in po-
ular, bridge dimensions, number of piers, material, tential network disruption. The importance of long-
design principle, foundation type are the main param- term affects should be defined based on local trans-
eters; the availability and accuracy of data influence portation needs.
the modelling. In either case, a bridge failure has an impact on the
The analysis of the hydrodynamic forces that rep- overall performance of the road network, since a crit-
resent the load on the bridge structure is evaluated us- ical link has been taken out of service. Network data
ing computational fluid dynamics (CFD) techniques. are collected from publicly available sources such as
The C++ toolbox OpenFOAM is the adopted soft- Digimap® (https://digimap.edina.ac.uk/), which pro-
ware, being open-source and particularly versatile for vides Ordnance Survey road maps. These contain
the development of customized numerical solvers
topographic information including name, location, allows for localized loads to be measured on individ-
length, capacity, type of road of transport networks. ual faces of a structure, which may be used to deter-
The last step is to adopt a transport model to under- mine whether or not individual components fail ver-
stand the impact of bridge failure on the wider sus entire structures.
transport network. Given the complexity of the study, For this study, the three-dimensional CFD code
the macro-transport model ESRI™ ArcGIS Network OpenFOAM was selected. Flood flows were mod-
Analyst (https://bit.ly/2GPMknl) is chosen for its ease elled using the interFoam solver, which is a two-
of use and visualisation. After configuring the trans- phase solver that relies upon Volume of Fluid (VoF)
portation network model with the collected data, rout- method (Tryggvason et al., 2011) to track the inter-
face between water and air phases. The underlying
ing and accessibility problems can be investigated us-
governing equations that are implemented in inter-
ing network-based spatial analysis. This analysis Foam are the Reynolds-averaged Navier-Stokes
links the individual failure of a bridge with the local (RANS) equations, which are solved using a predic-
road network the bridge serves for; thus, this frame- tor-corrector or projection type of method to solve for
work ultimately assesses the systemic impact of velocity and pressure fields, and advection equations
floods to the transport network due to bridge failure. for the volume fraction introduced by the VoF
method. More specifically, pressure-velocity cou-
2.1 Fluvial flooding with LISFLOOD-FP model pling was achieved using the PIMPLE algorithm,
which is a combination of the pressure-implicit split-
LISFLOOD-FP is a two-dimensional, spatially operator (PISO) and semi-implicit method for pres-
distributed, grid-based hydrodynamic model for sim- sure-linked equations (SIMPLE). Since the RANS
ulating channel and floodplain flows. The model dy- system of equations does not constitute a well-posed
namically simulates flood propagation in each grid system due to the so-called Reynolds stress tensor
cell at each time step, on the basis of the local inertial that arises from the Reynolds-averaging process, a
formulation of the shallow water equations and an ex- suitable turbulence model that introduces additional
plicit finite difference method. Numerically, this in- equations must be chosen to close the system. For this
volves calculating the momentum equation (the flow study, the k-ω Shear Stress Transport (SST) model
between cells given the mass in each cell) and the was used due to its ability to handle severely-sepa-
continuity equation (the change in mass in each cell rated flows near sharp corners better than other simi-
given the flows between cells) (Neal et al., 2018). The lar models such as the Standard, Renormalization
equations underpinning the model, including their Group (RNG), or Realizable k-ε models.
derivation, can be found in Bates et al. (2010) and de
Almeida et al. (2012).
As input data, LISFLOOD-FP needs a DEM of the 2.3 Structural analysis
area, channel and boundary condition information Finite-element (FE) analysis is commonly em-
(e.g. channel friction, width and depth, hydrograph) ployed in the structural engineering community to
and evaporation). Flow depth and velocity (for each simulate the response of bridges to natural hazards.
cell) are the output considered, since they are the in- Modern bridge superstructures are commonly formed
tensity measures of the hazard adopted by this study. of girders, cap beams, and columns which can be
The impact of bridges on flow is not explicitly repre- modeled accurately as line elements. Such an ap-
sented in this particular application. proach has been frequently used to model bridges in
OpenSees (McKenna et al. 2010) due to its nonlinear
2.2 Computational fluid dynamics (CFD) modeling capabilities; thus, this software is employed
in this study.
Compared to two-dimensional, depth-averaged OpenSees is seldom used to model structural re-
software such as LISFLOOD-LP, three-dimensional sponse to fluids, but Stephens et al. (2017) demon-
computational fluid dynamics (CFD) software is ca- strated how it can be loosely coupled with Open-
pable of resolving fine details of flood flow around FOAM to characterize structural response due to
bridges on a local scale such as splashes, eddies, or sequential earthquake and tsunami loading. A similar
flow separation, which cannot be captured by depth- loosely coupled scheme is used in the present work,
averaged methods. Also, bridges present a problem where (1) the bridge deck and girders are modeled as
for depth-averaged tools since the computational a rigid cross section (i.e., in 2D) and subjected to
mesh is two-dimensional and cannot be discretized flood flows at different water depths and velocities in
vertically, which does not allow for a gap underneath OpenFOAM, (2) the steady-state reactions on the
a bridge superstructure. Accurately modelling such cross section are recorded, and (3) the steady-state re-
behaviours is crucial when estimating flow-induced actions are applied as distributed external loads on
force demands, which requires the use of a fine, three- girder line elements in a 3D OpenSees model which
dimensional mesh. Additionally, using such a method represents the bridge superstructure. This OpenSees
model is used to evaluate structural limit states such
as unseating of the bridge deck; yielding of the gird- The OpenSees model was developed using line el-
ers, cap beams, or columns; or other limit states asso- ements for the reinforced-concrete columns and pre-
ciated with large enough deformation to invalidate the flex girders (a form of prestressed, concrete-encased
assumed configuration of the CFD model. steel beams). The girders had linear-elastic transla-
tional springs at each end to represent rubber
beari/ngs. These bearing elements were connected to
3 APPLICATION AND RESULTS rigid links which simulated cap beams, providing a
load path between the girders and columns. The
Carlisle is a flood-prone city (2011 population: OpenFOAM reaction forces and roll moments were
75,306) located in North West England (UK) (Fig. 3). applied as distributed loads in OpenSees on each
Three road bridges connect the two parts of the town bridge girder (i.e., over all 8 spans with 20 girders per
over the river Eden from North to South: the A689, span). Note that gravity loading was applied prior to
A7 and M6 bridges. As a proof of concept, the M6 the hydraulic loads.
highway bridge over the river Eden was considered. The primary structural limit state investigated was
A schematic model of this bridge is shown in Fig. 4 unseating of the girders. This was measured using the
with approximate column (reinforced concrete) and vertical force in each bearing. Figure 5 shows the dis-
girder (preflex beam) dimensions. tribution of vertical forces (Nf) normalized by weight
(Ng) for a 15-m water depth with a 3 m/s flow veloc-
ity. An Nf/Ng ratio of 1.0 indicates decompression of
the bearing and, consequently, exceedance of the un-
seating limit state.

Figure 5. Bearing vertical forces for 15-m depth and 3-m/s

flow velocity (positive indicates uplift).
Figure 3. The city of Carlisle is connected North-South by
three road bridges: the A689, A7 and M6 (highway) bridges. Under the range of loading investigated, this limit
state was not exceeded. Figure 6 shows the variation
of maximum bearing vertical force with water depth
and velocity. Similarly, Fig. 7 shows the variation of
maximum bearing shear force with water depth and
velocity. These results indicate that (1) the vertical
force on the bearings reaches a maximum at different
depths depending on velocity but tends to stabilize af-
ter the bridge is fully inundated and (2) the shear force
on the bearings increases approximately linearly with
depth and velocity even after full inundation.

Figure 4. Approximate geometry of M6 bridge with column

and girder sections shown (sections not to scale).

LISFLOOD-LP was set up to simulate flooding

scenarios for a domain covering 14.75 km2 of Car-
lisle, at 5m of resolution, for a flooding scenario with
the return period of 1-in-500-year.
The OpenFOAM model was set to simulate a wide
range of flood velocities and depths: steady-state ve-
locities of 1 through 3 meters per second and inunda-
tion heights of 15 meters to 18 meters were consid-
ered. The model measured forces on 20 individual
components along the cross-section of the bridge cor-
responding to girders and each girders’ approximate
tributary area.
 4 DISCUSSION AND CONCLUSION
Figure 6. Variation of maximum uplift force on bearings with
depth and velocity (positive indicate uplift). This study developed an integrated method that cou-
ple practices of computational flood dynamics (CFD)
The accessibility for the city centre of Carlisle is with reliability and network analysis. For the city of
evaluated for the baseline, i.e. non-perturbed condi- Carlisle (UK), a 1-in-500-years flooding event was
tions (Fig. 8), in terms of serviced area. This area con- simulated and the resulting hydrodynamic forces on
nects equal travel time location from the city centre. the highway bridge (M6) simulated. While simulated
A travel time of 8 minutes (480 seconds) is assumed, hydrodynamic forces and finite-element (FE) analy-
to respect the statutory 8-min threshold for response sis did not show uplift failure, overtopping of the
times of emergency services, such as ambulances. bridge is shown to occur at inundation heights of 14
The baseline serviced area measures 118.4 km2 . If the m and above. Given the potential for flood-water dis-
M6 bridge fails, the serviced area reduces to 116.0 ruption of traffic, this should be considered temporary
km2; the reduction is less impactful than expected be- network failure in its own right. For this particular lo-
cause other bridges in the North of town are supposed cation, the limit state was not hit, however the bridge
to be still in operation (A7 and A689 bridges). functionally failed (i.e. was not fit for purpose), even
However, the obtained result is a “best case sce- if the structural did not.
nario” since traffic is not accounted for, therefore the
likely congestion on the alternative routes is not con- The importance of this study consists in the proof of
sidered. Moreover, the presence of floodwater on the concept of a new holistic methodology and will be
streets is considered due to riverine flooding only, further developed by looking at a wider range of
thus surface water flooding is ignored. The impact of structures. A two-dimensional network model can
surface water flooding is potentially relevant, since give relevant community leaders a good starting point
the flood simulation domain is smaller than the ser- for susceptibility to flooding in the vicinity of a
viced areas. bridge. Because of the critical nature of bridges to a
transportation network, however, higher fidelity
modeling to better capture the engineering response
of lifelines within the community. This work presents
conceptual results using a combined CFD-FE ap-
proach to improve the fidelity of the network failure
predictions. By using a high-fidelity 3D analysis ap-
proach, designers can better predict the flow in the vi-
cinity of the bridge while including 3D effects (e.g.,
variations in the vertical dimension that include the
clearances under a bridge) and thus better predict lo-
cal fluid pressures that may lead to structural failure.
These forces can then be applied directly into a tradi-
tional FE model to better predict the global structural
Figure 7. Variation of maximum shear force on bearings response and corresponding critical structural compo-
with depth and velocity. nents. Ultimately this approach can be applied to any
coastal or riverine region where large-scale water in-
undation is expected.
Figure 8. Serviced areas for 8-minute travelling time in Future work should investigate the impacts of
baseline (non-disrupted) conditions and when the M6 bridge is other limit states (e.g., scour) which could result in
disrupted. bridge closure. Such analyses would benefit from 3D
CFD and finite-element models to help refine de- Kerenyi K., Sofu T. and Guo, J. (2009). Hydrodynamic
mands on the structure and reduce uncertainty in the forces on inundated bridge decks, Federal Highway Admin-
predicted bridge reliability. istration, FHWA-HRT-09-028.
Next steps of this study will analyze the impact of Liu L., Frangopol D.M., Mondoro A. and Yang D.Y.
the closure for each bridge and any combination of (2018). Sustainability-Informed Bridge Ranking under Scour
them. Further development could also involve the es- Based on Transportation Network Performance and Multiat-
timation of direct (e.g. repairs) and indirect costs (e.g. tribute Utility. Journal of Bridge Engineering, 23(10):
04018082. DOI: https://doi.org/10.1061/(ASCE)BE.1943-
rerouting) due to bridge closure.
5592.0001296
Lomonaco P., Alam M. S., Arduino P., Barbosa A., Cox
D.T., Do T., Eberhard M., Motley M.R., Shekhar K.,
ACKNOWLEDGEMENTS Tomiczek T., Park H., van de Lindt J.W., Winter A. (2018).
Experimental modeling of wave forces and hydrodynamics on
This work was supported by the Engineering and elevated coastal structures subject to waves, surge or tsunamis:
Physical Sciences Research Council (ESPRC) LWEC the effect of breaking, shielding and debris. Coastal Eng. Pro-
(Living With Environmental Change) Fellowship ceedings 1 (36): 53. DOI:
(EP/R00742X/1 and 2). https://doi.org/10.9753/icce.v36.waves.53
McKenna, F., Scott, M.H., and Fenves, G.L. (2010). Non-
linear finite-element analysis software architecture using ob-
REFERENCES ject composition. J. Comput. Civ. Eng. 24: 95-107.
Mondoro A. and Frangopol D.M. (2018). Risk-based cost-
AASHTO (2002). Standard specifications for highway benefit analysis for the retrofit of bridges exposed to extreme
bridges, 7th Ed., Washington, DC. hydrologic events considering multiple failure modes. Eng.
Arrighi C., Pregnolato M., Dawson R., Castelli F. (2019). Struct., 159: 310-319. DOI: https://doi.org/10.1016/j.eng-
Preparedness against mobility disruption by floods. Science of struct.2017.12.029
the Total Env., 654: 1010-1022. DOI: Mondoro A., Frangopol D.M. and Liu L. (2019). Bridge
https://doi.org/10.1016/j.scitotenv.2018.11.191 Adaptation and Management under Climate Change Uncer-
Bates P.D., Horritt M.S. and Fewtrell T.J. (2010). A simple tainties: A Review. Natural Hazards Review, 19(1): 04017023.
inertial formulation of the shallow water equations for efficient DOI: 10.1061/(ASCE)NH.1527-6996.0000270
two-dimensional flood inundation modelling. J. Hydrol., Motley M.R., Wong H.K., Qin X., Winter A.O. and Eber-
387(1–2): 33-45. DOI: 10.1016/j.jhydrol.2010.03.027 hard M.O. (2016). Tsunami-induced forces on skewed bridges.
Carey T.J., Mason H.B., Barbosa A.R., Michael H.S. J. Waterway, Port, Coastal, Ocean Eng. 142(3): 04015025.
(2019). Multihazard Earthquake and Tsunami Effects on Soil– https://doi.org/10.1061/(ASCE)WW.1943-5460.0000328
Foundation–Bridge Systems. J. Bridge Eng., 24(4): 04019004. Neal J.C., Bates P.D., Fewtrell T.J., Hunter N.M., Wilson
DOI: https://doi.org/10.1061/(ASCE)BE.1943-5592.0001353 M.D. and Horritt M.S. (2009). Distributed whole city water
Stephens, M.T., Winter, A., Motley, M.R., and Lehman, level measurements from the Carlisle 2005 urban flood event
D.E. (2017). Comparing seismic and tsunami load demands on and comparison with hydraulic model simulations. J. of Hy-
reinforced concrete and concrete filled steel tube bridges. Proc. drology 368(1-4): 42-55. DOI: https://doi.org/10.1016/j.jhy-
of the 39th IABSE Symposium. drol.2009.01.026
de Almeida G.A.M., Bates P.D., Freer J.E., Souvignet M. Neal J.C., Bates P.D., Fewtrell T.J., Hunter N.M., Wilson
(2012). Improving the stability of a simple formulation of the M.D., Horritt M.S. (2009). Distributed whole city water level
shallow water equations for 2-D flood modelling. Water Re- measurements from the Carlisle 2005 urban flood event and
sour. Res., 48(5): W05528. DOI: 10.1029/2011wr011570 comparison with hydraulic model simulations. J. Hydrol., 368
Dijkstra E.W. (1959). A note on two problems in connex- (1–4): 42-55. DOI: https://doi.org/10.1016/j.jhy-
ion with graphs. Numerische Mathematik, 1(1): 269-271. drol.2009.01.026
Ertugay K., Argyroudis S. and Düzgün H.Ş. (2016). Ac- Neal J., Dunne T., Sampson C., Smith A., Bates P.D.
cessibility modeling in earthquake case considering road clo- (2018). Optimisation of the two-dimensional hydraulic model
sure probabilities: a case study of health and shelter service LISFLOOD-LP for CPU architecture. Environ. Model. Softw.,
accessibility in Thessaloniki, Greece. Int. J. of Disaster Risk 107: 148-157. DOI: https://doi.org/10.1016/j.en-
Reduction, 17: 49–66. DOI: 10.1016/j.ijdrr.2016.03.005 vsoft.2018.05.011
Gidaris I., Padgett J. E., Barbosa A. R., Chen S., Cox D. T., Oudenbroek K., Naderi N., Bricker J.D., Yang Y., Van der
Webb B. and Cerato A. (2017). Multiple-hazard fragility and Veen C., Uijttewaal W., Moriguchi S. and Jonkman S.N.
restoration models of highway bridges for regional risk and re- (2018). Hydrodynamic and Debris-Damming Failure of
silience assessment in the United States: State-of-the-art re- Bridge Decks and Piers in Steady Flow. Geosciences, 8 (11),
view. J. Struct. Eng. 143 (3): 04016188. DOI: 409. DOI: https://doi.org/10.3390/geosciences8110409
https://doi.org/10.1061/(ASCE)ST.1943-541X.0001672 Ortúzar J.d.D. and Willumsen L.G. (2011) Modelling
Green D., Yu D., Pattison I., Wilby R., Bosher L., Patel R., Transport. Wiley.
Thompson P., Trowell K., Draycon J., Halse M., Yang L. and Pregnolato M., Sarhosis V., Kilsby C. (2018). Towards in-
Ryley T. (2017). City-scale accessibility of emergency re- tegrating modelling of flood-induced bridge failures. EPiC Se-
sponders operating during flood events. Nat. Hazards Earth ries in Engineering 3, 1698-1702. DOI: doi.org/10.29007/zthz
Sci., 17(1): 1-16. DOI: 10.5194/nhess-17-1-2017 Pregnolato M. (2017). Risk analysis of the disruption to
urban transport networks from pluvial flooding. PhD Thesis,
Newcastle University: Newcastle. Available at:
https://bit.ly/2SxtvbF
Pregnolato, M., Ford, A., Glenis, V., Wilkinson, S. and
Dawson, R. (2017a). “Impact of flooding and urban adaptation
in a changing climate”. J. of Infr. Sys., 23(4): 1-13. DOI:
10.1061/(ASCE)IS.1943-555X.0000372
Pregnolato M., Ford A., Wilkinson S. and Dawson R.J.
(2017b). The impact of flooding on urban transport sector - an
integrated depth-disruption function. Transp. Res. Part D:
Transport and Environment, 55: 67-81. DOI:
doi.org/10.1016/j.trd.2017.06.020
Pregnolato M., Ford A., Robson C., Glenis V., Barr, S. and
Dawson R.J. (2016). Assessing Urban Strategies for Reducing
the Impacts of extreme Weather on Infrastructure Networks.
Royal Soc. Open Sci., 3(5): 1-15. DOI: 10.1098/rsos.160023
Qin X., Motley M.R. and Marafi N. (2018). Three-dimen-
sional modeling of tsunami forces on coastal communities.
Coast. Eng., 140: 43–59. DOI:
https://doi.org/10.1016/j.coastaleng.2018.06.008
Tryggvason, G., Scardovelli, R. and Zaleski, S. (2011). Direct
numerical simulations of gas–liquid multiphase flows. Cam-
bridge University Press.
Turner D. (2016). Fragility assessment of bridge superstruc-
tures under hydrodynamic forces. MSc Thesis, Colorado State
Univ., Fort Collins, CO.
Yilmaz T., Banerjee S. and Johnson P. A. (2016). Perfor-
mance of two real-life California bridges under regional natural
hazards. J. Bridge. Eng. 21(3): 1–15. DOI:
https://doi.org/10.1061/(ASCE)BE.1943-5592.0000827
Wardhana K. and Hadipriono F. C., 2003. Analysis of Recent
Bridge Failures in the United States. J. Perf. of Constructed Fa-
cilities 17(3): 144–150. DOI:
https://doi.org/10.1061/(ASCE)0887-3828(2003)17:3(144)
Sun W., Bocchini P. and Davison B.D. (2018). Resilience
metrics and measurement methods for transportation infrastruc-
ture: the state of the art. Sustainable and Resilient Infrastructure,
DOI: 10.1080/23789689.2018.1448663
Winter A.O., Motley M.R. and Eberhard M.O. (2017). Tsu-
nami-like wave loading of individual bridge components. J. of
Bridge Eng. 23 (2), 04017137. DOI:
https://doi.org/10.1061/(ASCE)BE.1943-5592.0001177