dc.description.abstract |
The impact of natural hazards on communities, from a structural, environmental, social, and
economic perspective, is increasing worldwide. Following natural disasters, such as flooding,
earthquakes, tsunami, hurricanes, and volcanoes, transportation systems are disrupted resulting
in cities, towns, and villages being isolated for a period of time causing post-disaster response
issues, evacuation difficulties, accessibilities issues, increased travel costs, and economic
losses.
New Zealand (NZ) is at risk from different types of hazards and has experienced a variety of
disasters in recent years. The source of most disasters in NZ are geological hazards
(earthquakes, volcanoes, erosion, and landslide); and meteorological hazards (wind and heavy
rainfall). Since 2000, three major earthquakes have occurred in NZ, all in the South Island,
causing 184 deaths and an estimated economic loss of US$28.5 billion in damage. The Alpine
fault, one of the major fault systems in NZ, extends over 600km in the South Island. There is a
high probability of a rupture in the next 50 years, with resulting severe damage to the built
environment, especially to infrastructure, predicted. In order to assess the operational
performance and resilience of the South Island regional road network following a potential
Alpine Fault Magnitude 8 earthquake, this project was funded by the Ministry of Business,
Innovation, and Employment (MBIE) under the National Science Challenge (NSC) –
Resilience to Nature’s Challenges (RNC) program. Although the physical impact of the AF8
has been studied in detail, there is a paucity of research in the literature assessing the operational
performance of the road network following such an event. Only one study has estimated the
operational performance of the road network post-disaster without using any transportation simulation model. The main contribution of this research is, therefore, to develop, calibrate and
validate a mesoscopic model of the South Island road network and develop a novel approach to
assess the resilience of the network following a disaster.
To assess the operational resilience of transportation networks following a natural disaster,
three approaches have been reported in the literature, namely: conceptual, analytical and a
combination of analytical and simulation. A combination of analytical and simulation was
chosen for this research.
In order to develop the simulation model, the road network from Open Street Map was first
imported into the software. The South Island was divided into 541 traffic zones based on NZ
geographic unit areas based on census data. An Origin-Destination (OD) matrix was created for
light vehicles (commuting and tourism trips) and heavy vehicles (freight trips) based on 2013
census data, land use GIS shapefiles, Regional Tourism Organisation data, and Ministry of
Transport data. The model was first calibrated macroscopically against 2013 traffic data as the
base year by adjusting the OD matrix separately for light vehicles and heavy vehicles using the
matrix estimation and static assignment methods. Then the model was calibrated at the
mesoscopic level using the C-Logit based Stochastic Route Choice method. To validate the
model in a post-disaster environment, 7-day Average Daily Traffic (ADT) data starting from
Day 8 after the 2016 Kaikoura earthquake was applied. Corridor and trip analyses were
conducted to assess post-disaster operational performance of the road network. Corridor
analysis results indicate a significant increase in traffic count and density, and a minor decrease
in average travel speeds on four main corridors namely: SH65, SH63, SH6 (between SH63 and
SH65), and SH7 (between SH65 and SH1), serving as the main alternative routes after the
earthquake. Trip analysis results show a significant increase in the average travel time from Marlborough to other affected traffic zones (typically 20-50% and as much as 90% in the worst
case) due to the increased travel distances on the alternate routes.
The calibrated and validated model, therefore, was utilised to estimate the performance of the
road network impacted by the Alpine Fault Magnitude 8 earthquake scenario. The trip analysis
of one day, one week, six months, and beyond six months after the earthquake indicates that
around 2.02%, 1.16%, 0.39%, and 0.13% of total trips, respectively, cannot occur in the South
Island due to accessibility issues. Almost all of the inter-district trips from the three main
impacted districts, namely Buller, Westland, and Grey, would be cancelled for at least one week
after the earthquake.
To assess the resilience of road network following a disaster, a novel Normalised Trip
Resilience (NTR) measure was proposed incorporating three different dimensions of resilient
systems, namely; robustness, redundancy, and recovery. The proposed measure reflects the
resilience of trips in percentage for the whole of the recovery period. The Equivalent number
of Impacted Trips (EIT) is proposed as a measure of effectiveness to prioritize the importance
of OD pairs post-disaster using NTR and flow. The proposed measures were implemented to
evaluate the operational performance of New Zealand’s South Island road network in the event
of the Alpine Fault Magnitude 8 earthquake scenario. The resulting measures were capable of
being calculated from the outputs produced by a transportation simulation model thereby
verifying their practicality in real-world situations. The importance of including both robustness
(represented by the number of eliminated trips) and redundancy (represented by increased travel
time), over the horizon of the post-disaster recovery phase was highlighted. Eliminated trips
contributed significantly in areas that were cut off and isolated post-disaster, due to a lack of
alternative routes, and increased travel time contributed as more roads were reopened but the
alternative routes resulted in increased travel distances and, consequently, travel time. |
|