3.2.2. Impacts on Traffic in Bristol

Within the Bristol model, as we also considered the temporal aspect of the flooding in the analyses such that we can specify the time and duration of the flood event. In this example we have specified that flooding occurs as 7 am and lasts for 30 min. During this period, the maximum permitted speeds along hazard affected road sections are temporarily modified and will return to normal at 7:30 am after the flood event has ended.

As the Bristol model is micro-scale, if some journey starts within a flooded region they cannot begin and therefore, are not added to the network this can lead to reduced traffic number on the road during and after the flood event and needs to be considered as part of the impact assessment. Table 8 shows the percentage of journeys whose start-times begin during a flood event and are unable to begin their journey as lie within a closed road.


**Table 8.** Journeys unable to begin due to flooding.

One of the additional indicators used to assess the impacts of traffic flows through a micro-scale simulation is to examine the number of vehicles within the network at any given time under flooded conditions and compare this distribution against dry weather conditions. Under flooded conditions, as some road sections will temporarily have their maximum allowable speeds reduced and some sections are temporarily closed. The journey times for vehicles that usually traverse these sections along their assigned routes will increase as vehicles are forced to either move at a reduced speed through shallow water or are diverted onto alternative routes if their original route is obstructed. Figure 15 shows a comparison of the number of vehicles within the road network over time for the different severities of flood events for baseline and future climate change conditions. Within this figure, the "Dry Count Range" represents the range (minimum and maximum) of vehicle counts across the 10 generated OD matrix routing scenarios, and the "Flooded Traffic Count Range" shows the ranges with respect to the network during the 30 min flood simulations. The "Average Flood Traffic Count" represents the average number of vehicles within the network during the respective flood scenarios. The figures highlight that even though a percentage of journeys are lost/unable to start during the flood event, the vehicular saturation of the network both during and immediately after the flood event surpasses the dry weather conditions with the network (on average) recovering by 9 am for all scenarios. The Flooded Traffic Count Range further highlights the different effects flooding has across the 10 generated route scenarios. Figure 16 further shows that even after the flood event has finished the road network still takes time to recover as previously impeded vehicles are continuing to complete their journeys and their remaining presence within the network effects other vehicles that travelling.

Figure 17 shows the comparison of the relative EP curves for the Baseline and Future Climate Change scenarios utilising the same cost to speed relationship applied to the Barcelona case study. Here the points for the Baseline and Climate Change scenarios represent the average calculated losses derived from the simulations with the curves interpolated from these points respectively. Within this example, we are examining the relative cost increases between the hours of 7 am and 9 am that corresponds to the period of disruption shown in Figure 16. In contrast to the Barcelona case study the calculated loss values depicted for Bristol simulations are considerably less. There are a number reasons for this including, but not limited to, the case study area examined within the city of Bristol (24 km2) is considerably smaller than that assessed within Barcelona (102 km2). A second reason relates to the limited number of vehicles used in the duration of the model. With Bristol having a population of approximately 463,500 [25] and 41% of the population driving a car to work [26] the simulated

5000 vehicles over a 6 h period could be a significant under estimate of the traffic volumes/journeys undertaken within the network during this period. The results, therefore, merely serve to show how the effects of climate change can result in observed increases in disruption to traffic flows and potential losses within the traffic network.

**Figure 17.** Flood Impact on Traffic Relative EP Curves for Bristol.
