How does increasing impervious surfaces affect urban flooding in response to climate variability?. (November 2020)

• Record Type: Journal Article
• Authors: Sohn, Wonmin; Kim, Jun-Hyun; Li, Ming-Han; Brown, Robert D.; Jaber, Fouad H.
• Is Part Of: Ecological indicators. Volume 118(2020)
• Publication Date: 2020-11
• Subjects: Total impervious area -- Directly connected impervious area -- Stormwater management -- Rainfall characteristics -- Climate variability -- Ordinal logit regression; Environmental monitoring -- Periodicals; Environmental management -- Periodicals; Environmental impact analysis -- Periodicals; Environmental risk assessment -- Periodicals; Sustainable development -- Periodicals; 333.71405
• DOI: 10.1016/j.ecolind.2020.106774
• Languages: English
• ISSNs: 1470-160X

Abstract

Highlights: TIA and DCIA positively affect the probability of runoff yield. DCIA contributes far more than TIA to generating far-above-average runoff depth. The contributions of TIA and DCIA to far-above-average peak flow are comparable. The impacts of TIA and DCIA are subject to climate variability. TIA and DCIA control is effective within a specific range of rainfall depth. Abstract: Total impervious area (TIA) is one of the most common measures for predicting runoff yield in hydrologic studies and regulating urbanization in land use policy. Directly connected impervious area (DCIA), a subset of TIA, represents the hydraulic connection between development and underground sewer systems. Which indicator to use in runoff prediction has been subject to debate. The effectiveness of TIA and DCIA in the face of climate variability also remains unclear. The present study empirically assessed the impacts of TIA and DCIA on urban runoff in three metropolitan statistical areas (MSAs) in the U.S. state of Texas. The total monthly runoff depths of 92 watersheds and peak flows of 43 watersheds were computed using streamflow monitored by the U.S. Geological Survey from 2010 to 2017. A series of ordinal logit regression models were developed to determine which imperviousness indicator better predicted probable runoff yield. Additionally, an average marginal effect analysis was performed to investigate how TIA and DCIA responded to changing precipitation depths. The results demonstrate that DCIA outperformed TIA in runoff depth prediction, whereas TIA predicted peak flow better than DCIA. However, for far-above-average runoff, the contribution of DCIA to runoff depth was far greater than that of TIA, but no difference was found for peak flow. The effectiveness of TIA and DCIA also varied by total and 24-hour peak depths of monthly precipitation. After reaching their maximum capacity, both TIA and DCIA became less effective in predicting runoff and did not correlate with rainfall depth in extremely wet months. Meanwhile, the control of TIA and DCIA for runoff volume reduction was most effective for monthly rainfall of a 5% to 10% probability of exceedance in all MSAs, whereas that of peak flow reduction was most effective if the 24-hour peak storm in a month had a 2% to 5% probability of exceedance. The findings of this study demonstrate the hydrologic significance of regulating DCIA over TIA for high-risk runoff under certain rainfall depths and return periods. The study expands the current knowledge of urban hydrology for effective stormwater management and mitigation of future flooding risk.