Enhancing Urban Resilience to Surface Water Flooding: A Novel Approach Using UAS-Derived Topographic Indices

With the increase in urbanisation and climate change around the globe, there is an increased risk of surface water flooding. Although extreme flood events are commonly discussed, smaller, more frequent flood events also cause significant disruptions that impact human life and put financial stress on authorities. The majority of urban flooding is due to drainage failure. For effective surface water management, it is important to assess the effectiveness of existing surface drainage assets and accordingly plan asset maintenance or retrofitting of new drainage assets (both traditional and nature-based solutions). The surface drains usually fail because they are either not positioned where the surface water accumulates or are blocked and not maintained to meet the standards. Microtopography significantly influences the surface water flow movement, flow path, flow direction, and velocity, and consequently, dictates the areas of water accumulation.  Thefore, this study explores a novel approach to evaluate storm drain inlet positions using high-resolution topographic indices maps derived from Unmanned Aerial System (UAS) imagery. The Topographic Wetness Index (TWI) and Topographic Control Index (TCI) were employed to identify drains misaligned with surface water pathways and pinpoint critical drains in the sink points of the topography, respectively. 

Storm drain inlets were classified as functional or non-functional based on their intersection with the flow path defined by the optimal TWI threshold. The optimal threshold was determined to be the 90th percentile at a value of 6.19 based on the spatial similarity of the delineated runoff-contributing flow path with the 1 in 100 year surface water flood map produced by the Environment Agency. The validation of the classification of storm drains effectiveness based on TWI using field data yielded an overall accuracy of 53 %, 75% precision, and an F1 score of 62%, indicating a moderate success of TWI in identifying functional drains. Although validation with LIDAR data showed a slight improvement in accuracy and precision, the results generally demonstrated that TWI has a strong capability to correctly identify functional drains; however, it is slightly more challenging to identify nonfunctional drains. 

A comparison of the UAS-derived TCI map with the LIDAR-derived TCI map demonstrated a 90% match in the identified sink areas and a high accuracy of 93% in identifying critical drains in the sink areas. The results suggest that the combined use of TWI and TCI offers a promising approach for assessing storm drain effectiveness, based on its position and guiding authorities in identifying areas with drainage deficits and preparing targeted drainage maintenance strategies. The findings of this research provide valuable insights for urban planners and decision-makers to not only optimise the placement and maintenance of storm drain inlets but also highlight the potential for alternative nature-based low-impact development (LID) solutions in locations where traditional drainage is found to be inefficient. This would ultimately enhance the resilience of urban areas to surface-water flooding.