Quantifying the effects of stream channels on storm water quality in a semi-arid urban environment

Summary

Stormwater drainage systems can have a large effect on urban runoff quality, but it is unclear how ephemeral urban streams alter runoff hydrochemistry. This problem is particularly relevant in semi-arid regions, where urban storm runoff is considered a renewable water resource. Here we address the question: how do stream channels alter urban runoff hydrochemistry? We collected synoptic stormwater samples during three rainfall–runoff events from nine ephemeral streams reaches (three concrete or metal, three grass, three gravel) in Tucson, Arizona. We identified patterns of temporal and spatial (longitudinal) variability in concentrations of conservative (chloride and isotopes of water) and reactive solutes (inorganic-N, soluble reactive phosphorous, sulfate-S, dissolved organic carbon (DOC) and nitrogen, and fecal indicator bacteria). Water isotopes and chloride (Cl) concentrations indicate that solute flushing and evapoconcentration alter temporal patterns in runoff hydrochemistry, but not spatial hydrochemical responses. Solute concentrations and stream channel solute sourcing and retention during runoff were significantly more variable at the grass reaches (CV = 2.3 − 144%) than at the concrete or metal (CV = 1.6 − 107%) or gravel reaches (CV = 1.9 − 60%), which functioned like flow-through systems. Stream channel soil Cl and DOC decreased following a runoff event (Cl: 12.1–7.3 μg g⁻¹ soil; DOC: 87.7–30.1 μg g⁻¹ soil), while soil fecal indicator bacteria counts increased (55–215 CFU g⁻¹ soil). Finding from this study suggest that the characteristics of the ephemeral stream channel substrate control biogeochemical reactions between runoff events, which alter stream channel soil solute stores and the hydrochemistry of subsequent runoff events.

Introduction

In water limited areas such as the semi-arid southwestern United States (US), urban runoff is a water resource that, through active management as focused recharge, can augment limited water supplies and enhance localized renewable groundwater resources (Hoffmann et al., 2007). Urbanization in arid and semi-arid regions can negatively impact stream and drinking water quality by increasing loads of nutrients, metals and organic pollutants to surface and ground water (Asaf et al., 2004, Barber et al., 2006, Carlson, 2004, Gallo et al., 2012; e.g. Hamilton et al., 2004, Lewis and Grimm, 2007), yet little is known regarding how the characteristics of the stormwater drainage system, and specifically the characteristics of the stream channel substrate, alter urban runoff hydrochemistry.

In the semi-arid Southwest, increasing imperviousness has been shown to enhance partitioning of rainfall into runoff, resulting in higher runoff ratios and more frequent and longer runoff duration (Gallo, 2011, Gallo et al., in review, Kennedy, 2007). However, the characteristics of the stormwater drainage system appear to be more important in controlling urban runoff hydrochemistry than catchment imperviousness (Gallo et al., in review, Grimm et al., 2005, Lewis and Grimm, 2007, Meierdiercks et al., 2010). It is likely that in semi-arid regions, increases in runoff frequency and duration promote ephemeral stream channel wetting and expand the spatial and temporal distribution of biogeochemical processes. Urban ephemeral channels may therefore act as biogeochemical hot spots, or regions of relatively high reaction rates compared to the surroundings (McClain et al., 2003), and alter hydrochemical responses by diminishing or immobilizing solutes in runoff.

Biogeochemical processes in streams are tightly linked to the characteristics of the channel substrate (e.g. Findlay et al., 2011). Despite extensive research assessing the effects of stormwater management strategies on urban runoff quality (e.g. Collins et al., 2010, Goonetilleke et al., 2011, Hatt et al., 2009, Weiss et al., 2007), information on how urban stream channels alter runoff hydrochemistry is limited. Field studies indicate that semi-arid urban and agricultural headwater streams have higher potential rates of denitrification than natural streams probably owing to higher stream nitrate concentrations and lower stream velocities (Findlay et al., 2011). In addition, variable redox conditions (Collins et al., 2010), the substrate’s hydraulic conductivity and organic matter content, as well as the presence of vegetation (Hatt et al., 2009) can alter solute sourcing and immobilization, which in turn alter urban runoff hydrochemistry.

However, processes unrelated to the characteristics of the stream channel, such as solute evapoconcentration (Henderson and Shuman, 2010, Meredith et al., 2009, Moravec et al., 2010), solute mobilization from the uplands (Lee et al., 2004, Schiff and Tiefenthaler, 2011) and water column biogeochemical processes like microbial respiration (Mallin et al., 2006) can alter stream water quality, but these factors are often overlooked in urban stream studies. Consequently, successful management of urban runoff requires understanding both water column processes and stream channel controls on urban runoff hydrochemistry.

In this paper, we address the role of the stream channel on modifying runoff quality by addressing the question: how do stream channels alter urban runoff hydrochemistry? We characterize patterns and potential controls of longitudinal urban runoff hydrochemistry in three distinct types of urban waterways: impervious (specifically concrete channels and metal pipes), gravel and grass lined stream channels. We use stable isotopes of water and conservative trace chloride (Cl) to identify occurrences of solute flushing, evapoconcentration, and conservative transport. During periods of conservative Cl transport, we evaluate the potential effects of the stream substrate in controlling runoff hydrochemistry in three types of stream reaches. Finally, we quantify changes in ephemeral channel soil solute stores following a rainfall–runoff event. This work is part of a larger study evaluating the impacts of urban land use on storm runoff and water quality across the Tucson Basin (Gallo et al., 2012, Gallo et al., in review) and took place during the summer rainfall season, when most streamflow occurs (Stone et al., 2008) and when biogeochemical processing during and between runoff events, may be of particular importance to urban runoff quality and stormwater management.