Evaluation of resilience of urban road networks to flood induced congestion in Greater Kampala Metropolitan Area

Abstract

Recent global natural and man-made evolutions have made urban road networks in transportation systems vulnerable to natural disasters, accidents, and infrastructure failures, among others. Such events disrupt smooth network traffic flow leading to heavy congestion as road users attempt to reroute from a link to alternative links to reach particular destinations. It is imperative to assess whether urban networks can continue to function effectively in the wake of these challenges. Previous studies provided methods of analyzing resilience of road networks using metrics like robustness, network connectivity, vulnerability index, redundancy, and recovery time. This study has evaluated the resilience to flood induced congestion of a selected zone urban road network using the network vulnerability approach. The flooding spot within the study zone is located along Banda – Kireka road in the eastern part of Kampala city. The objectives were: (1) to determine critical flood depth of a flooded spot on a link within this network, (2) to develop network models for normal traffic, partial closure and complete closure of the link with the flooded spot, and (3) to determine vulnerability of selected links in the studied zone so as to establish measures for enhancing network resilience to pluvial flooding. The study used previous rainfall data to develop a Digital Elevation Model for the catchment of the flooded spot, a Real-Time Kinematic machine to do sub-catchment topographic surveys, Manual classified counts for traffic data collection, analysis of existing Origin-Destination data, ArcGIS 10.2 to analyse the DEM. Catchment hydrology and channel hydraulics were used to estimate flooding at the flooded spot using Storm Water Management Model. Planung Transport Verkehr (PTV) VISUM was used to develop the road network, distributed traffic and assignment of the Origin - Destination matrix onto the road network. The results show that rainfall events of return period equal to or greater than 5 years cause significant flooding of the trouble spot leading to partial or complete closure. The models for the studied scenarios established that flooding significantly disrupts the performance of zonal networks in the GKMA, during partial and complete closure resulting in reduced traffic flow on affected links. These disruptions are driven by drivers' perceptions of congestion during rainy events, leading them to seek alternative routes to their desired destinations. Out of the selected 22 study links in the network, three specific links, namely Nakawa–Bugolobi, Spear–Kyambogo, and Katalima roads were found to be vulnerable for all the traffic flow scenarios. These critical links require urgent attention and capacity improvements to accommodate increased traffic flow during flood events and enhance their resilience. It is therefore important to apply such an approach to networks in other zones so as to identify vulnerable links and give them priority during resource allocation. Further research is needed to evaluate the impact of flooding on network performance during Post-Meridiem (PM) peak periods, to understand how flooding affects traffic flow throughout the day and to provide strategies for resilience enhancement to flooding.