On the effectiveness of green infrastructure to reduce stormflow at catchment scale

Population growth and climate change alter the urban water cycle resulting in increasing frequency and magnitude of urban floods. In this study, we compared stormwater response in an urban drainage system between two adjacent urban sewersheds in Buffalo, NY, USA. At the first site (DEL), comprehensive installations of green infrastructure (GI) (i.e. bioretention cells) were carried out, while the second site (SQ) was minimally influenced by GI practices. Stormflow was monitored as pipeflow at both sites for an observation period of five years, three pre-construction and two post-construction years. We identified storm events and calculated event runoff, as excess flow above baseflow. Additionally, we evaluated annual total flow and peakflow (annual and seasonal) between the sites and between pre- and post-construction. Our analyses were confined to snow-free seasons because storage of precipitation in the snowpack confounds the evaluation of the precipitation-runoff relation. The analysis showed that the GI implementation was highly effective in reducing stormflow. Total annual flow was reduced at DEL between pre- and post-construction, while no trend was observable at the minimally influenced by GI SQ. Also, event-based stormflow was reduced through GI implementation across all snow-free seasons. Last, median event peakflow was clearly reduced through GI, especially in spring and summer, whereas results during fall were less clear. Through this hydrometric analysis, this study is among the first that provided evidence for the efficiency of GI in reducing stormflow beyond the plot-scale and thus provides future guidance on flood mitigation in urban environments.