Evaluation of infiltration‐based stormwater management to restore hydrological processes in urban headwater streams

Abstract

Urbanization threatens headwater stream ecosystems globally. Watershed restoration practices, such as infiltration‐based stormwater management, are implemented to mitigate the detrimental effects of urbanization on aquatic ecosystems. However, their effectiveness for restoring hydrologic processes and watershed storage remains poorly understood. Our study used a comparative hydrology approach to quantify the effects of urban watershed restoration on watershed hydrologic function in headwater streams within the Coastal Plain of Maryland, USA. We selected 11 headwater streams that spanned an urbanization–restoration gradient (4 forested, 4 urban‐degraded, and 3 urban‐degraded) to evaluate changes in watershed hydrologic function from both urbanization and watershed restoration. Discrete discharge and continuous, high‐frequency rainfall‐stage monitoring were conducted in each watershed. These datasets were used to develop 6 hydrologic metrics describing changes in watershed storage, flowpath connectivity, or the resultant stream flow regime. The hydrological effects of urbanization were clearly observed in all metrics, but only 1 of the 3 restored watersheds exhibited partially restored hydrologic function. At this site, a larger minimum runoff threshold was observed relative to the urban‐degraded watersheds, suggesting enhanced infiltration of stormwater runoff within the restoration structure. However, baseflow in the stream draining this watershed remained low compared to the forested reference streams, suggesting that enhanced infiltration of stormwater runoff did not recharge subsurface storage zones contributing to stream baseflow. The highly variable responses among the 3 restored watersheds were likely due to the spatial heterogeneity of urban development, including the level of impervious cover and extent of the storm sewer network. This study yielded important knowledge on how restoration strategies, such as infiltration‐based stormwater management, modulated—or failed to modulate—hydrological processes affected by urbanization, which will help improve the design of future urban watershed management strategies. More broadly, we highlighted a multimetric approach that can be used to monitor the restoration of headwater stream ecosystems in disturbed landscapes.

Stormwater management network effectiveness and implications for urban watershed function: A critical review

“…Several empirical studies show reductions in peak flow from infiltration and harvest SCM networks (Bedan & Clausen, ; Jarden et al, ; Wilson et al, ), with one study documenting an order of magnitude decrease in median peak flow (Wilson et al, ). Run‐off thresholds, or the minimum rainfall depth required to generate run‐off, were higher in watersheds implemented with infiltration and harvest focused SCM networks than in watersheds with direct conveyance or detention‐based SCMs (Fanelli et al, ; Hood et al, ; Loperfido et al, ). However, in watersheds implemented with the same SCM design, run‐off thresholds decreased with greater impervious cover, suggesting a decreased SCM benefit in watersheds with high impervious cover (Fanelli et al, ).…”

“…In another small watershed, groundwater levels were found not to significantly increase over time as urbanization with infiltration facilities took place, although stormflow became better controlled (Kessler et al, ). Although an infiltration‐based SCM in Annapolis, Maryland, United States, did intercept run‐off for small rainfall events, baseflow in the stream was significantly lower than in forested reference streams; there was no difference in baseflow between the watershed implemented with SCMs and an adjacent urban watershed with no SCM implementation (Fanelli et al, ). Infiltration SCMs implemented in Boston were found to raise water tables in a small but significant way (Thomas & Vogel, ).…”

“…Percent impervious area in watersheds studied have ranged from 3.8% to 85% ( Figure 2). In a few cases, all impervious area is "treated" (drains to an SCM) (e.g., Fanelli, Prestegaard, & Palmer, 2017;Hogan, Jarnagin, Loperfido, & Van Ness, 2014). However, other studies have focused on watersheds with implementation of SCMs for only a fraction of the impervious area (e.g., Bell, McMillan, Clinton, & Jefferson, 2016;Jarden, Jefferson, & Grieser, 2016).…”

Hydrologic metrics for status‐and‐trends monitoring in urban and urbanizing watersheds

“…In contrast to the rich literature on discharge‐based hydrologic metrics, evaluation of stage‐based metrics is not readily identifiable in either agency reporting or formal publications. We therefore consider this a preliminary evaluation only, with the hope that it will spur further evaluation of whether the reduced cost of implementation improves the feasibility of collecting meaningful data (e.g., Fanelli, Prestegarrd, & Palmer, ).…”

Modelling hydrological response to a fully‐monitored urban bioretention cell

“…Broadly, base flow generation is sensitive to alterations in evapotranspiration and water table levels, which may not be fully restored by installation of GI (Bhaskar, Beesley, et al, 2016;Bhaskar, Hogan, & Archfield, 2016;Bhaskar, Welty, Maxwell, & Miller, 2015). As a possible consequence of variability arising in placement and effectiveness, urban areas with GI stormwater infiltration retrofits have been observed to have both higher (Bhaskar, Beesley, et al, 2016, Bhaskar, Hogan, et al, 2016 and lower (Fanelli, Prestegaard, & Palmer, 2017) base flow volumes compared to nearby forested catchments. At both small and large scales, soil water exfiltrating from a bioretention cell and encountering a lowerpermeability soil matrix may cause groundwater mounding to occur.…”