Abstract:
Flash flooding in the University of Alabama (UA) campus has become more frequent than before due to the rapid development throughout the campus and the City of Tuscaloosa. To function as a robust and flood free area, the UA campus requires updated and data-informed flood mitigation measures. Identifying flood-prone areas on the campus as well as the reason behind flooding in these areas is, therefore, an important first step before developing suitable environment-friendly solution. This research focuses on investigating key drivers and mechanisms controlling flooding in urban environments. The MIKE URBAN model - a 2D hydrodynamic framework is used to simulate the 6th July, 2018 flood event on the UA campus. The results show that ~7.7% area of the campus was flooded with maximum water depth of 0.78 m and volume 16,100 m3. Six out of seven major flood locations found in the campus shows the dominance of impervious surface ranging between 60-90%. Detail land cover classification in those flooded locations shows the presence of buildings (29%) and roads (23%) to be higher than other land covers. Following identification of flood-prone regions on campus, the flood-contributing factors are investigated through a field measurement campaign of infiltration rate, soil moisture, soil type, drainage system etc. The results of this analysis reveal that soil texture is quite homogeneous across campus (sandy loam) but with high degree of the spatial and temporal variation in infiltration rate and soil moisture. Comparison between different storm event return periods (1, 2, 10 and 100 years) and between the spatial resolution of the simulations (15, 10 and 5m) show the same flooding hot spots are persistent but with considerable variation in water depth and flood extent. To investigate the effect of stormwater infrastructure on flooding conditions, a simulation without the drainage infrastructure was conducted. The results show that stormwater infrastructure decrease flooding extent and volume by factors of 2.5 and 15 respectively. To identify the contribution of green space to flood mitigation, a comparison of two land cover simulations were compared to the realistic land cover simulation (entirely pervious, entirely impervious). The results show that actual and entirely impervious land cover increase flooding volume by factors of 2.6 and 3.6 respectively. Lastly, a combination of different green infrastructures BMPs has been proposed as potential flood mitigation measures.
