Green Infrastructure Recovery: Analysis of the Influence of Back-to-Back Rainfall Events
| Journal of Sustainable Water in the Built Environment | 2017 | Peer Reviewed | Original research | https://ascelibrary.org/doi/abs/10.1061/JSWBAY.0000819Abstract
A core concept in stormwater green infrastructure (GI) design is whether a system will meet its rainfall-runoff volume capture goals within a period of time after a previous event. In GI design, it is necessary not to view storms as singular, isolated events, but rather as a series of events, some of which are occurring within short durations and are often termed back-to-back events. This paper demonstrates the statistical rarity of back-to-back rainfall events that impact GI performance and analyzes the expected impact on the design of several GI systems for the mid-Atlantic region. Twenty-four scenarios were evaluated for common design events (2.5, 3.8, and 8.1 cm), followed by a substantial subsequent event (50-100% of the original storm volume) that occurred within the period where it would be expected that the GI system would be recovering capacity (24, 48, 72, or 96 h interevent period). The results indicated that only four scenarios had an annual average occurrence greater than one time per year, and 9 of the 24 scenarios had less than 0.1 annual average occurrences. Simple, conservative models of a bioinfiltration rain garden and green roof demonstrated that system storage capacity is almost always restored by infiltration and evapotranspiration within the prescribed interevent drawdown period, thus back-to-back events are not a primary concern. This finding was further confirmed with field site evidence from three GI sites at Villanova University, which regularly captured more than the design rainfall volume and only infrequently had minimal discharges for rainfalls smaller than design. An exception to this finding was that a green roof with a drainage layer did not exceed its design capacity because the drainage layer conveyed stored water away from the GI system before evapotranspiration could contribute to volume removal. These results demonstrate that the likelihood of large back-to-back events is very low, as is the chance of the design runoff volume being exceeded. This paper provides evidence that the existing required drawdown period for GI design can be overly restrictive for a system to meet its volume control goals, which may inhibit optimal implementation. The findings support more appropriate design through the development of regionally specific required drawdown times based on storm event frequency.