Determining Hydrologic Pathways of Streamflow Using Geochemical Tracers in a Claypan Watershed

Abstract

Despite the low permeability of claypan soils, groundwater has been heavily contaminated by nitrate in agricultural watersheds dominated by claypan soils. However, it is unclear how nitrate concentrations in groundwater affect stream water quality. In this study, streamflow pathways were investigated using natural geochemical tracers in the 73-km 2 Goodwater Creek Experimental Watershed in northeastern Missouri. Samples were collected from 2011 to 2017 from stream water (weekly-biweekly), precipitation (eventbased), groundwater in 25 wells with screened depths varying from 2 to 16 m (bimonthlyseasonal) and interflow above the claypan in 7 shallow piezometers (weekly-monthly). The results of endmember mixing analysis using major ions indicate that streamflow was dominated by near-surface runoff (59 ± 20%), followed by interflow (25 ± 16%) and groundwater (16 ± 13%). Analysis of endmember distances using the mixing space defined by stream water chemistry suggests that groundwater contributions to streamflow came primarily from the intermediate to deep glacial till aquifer near and below 8 m. Near-surface runoff was persistent and dominant even after isolated precipitation events during a prolonged dry period. It is hypothesised that the alluvial aquifer near stream banks acts as a mixing zone to receive and store various source waters, resulting in persistent delivery of runoff to the stream. Groundwater, even though its contribution was limited, plays a significant role in regulating streamflow NO 3 − concentrations. This study significantly improves our understanding of claypan hydrology and will lead to the development of models and decision support tools for implementation of management practices that improve groundwater and stream water quality in restrictive layer watersheds.

Regional Patterns and Physical Controls of Streamflow Generation Across the Conterminous United States

“…Insights from previous field investigations help to explain why low soil permeability leads to the dominance of SOF or SSF1. Previous field investigations demonstrated that the reported low saturated hydraulic conductivities in some regions (e.g., Till Plains and the Dissected Till Plains) might be attributed to the existence of a relatively impermeable soil layer (Liu et al., 2020; Zimmer & Gannon, 2018), for example, claypan which is a naturally formed argillic horizon (Bt). These impeding layers dampen deep percolation and favor the formation of a shallow and relatively permanent water table and generation of SOF or SSF1.…”

Controls on decadal, annual, and seasonal concentration‐discharge relationships in the Sleepers River Research Watershed, Vermont, northeastern United States

“…Calcium, Mg 2+ , and Na + were considered to be conservative, while K + was quasi-conservative. The relative rootmean square errors (RRMSE) were lower in 1-D and 2-D mixing spaces for Ca 2+ , Mg 2+ , and Na + than for other ions, but the value of SO 4 2À was not unusually high compared to other studies (e.g., Liu et al, 2013;Liu et al, 2020), suggesting that SO 4 2À could be considered to be quasi-conservative as well.…”

“…Second, it is easier to code the procedure from an operational perspective. Third, a low R 2 may still be significant at α = 0.001 with a large number of samples, as encountered in Liu et al (2020). A change was also made to validate EMMA results by simulating streamflow chemistry for conservative solutes.…”

“…As discussed in Supporting information, using conservative solute concentrations in either precipitation (for SOF) or shallow soil water (for NSR) yielded similar EMMA results in this study. As suggested by Liu et al (2020), NSR may become return flow which is SOF that occurs over part of watershed and/or certain time periods (Dunne & Black, 1970a, 1970b.…”

Streamflow generation in semi‐arid, glacier‐covered, montane catchments in the upper Shule River, Qilian Mountains, northeastern Tibetan plateau

“…Endmember contributions were resolved following Liu et al (2017) and Liu et al (2020). To estimate statistical uncertainties associated with spatiotemporal variation of tracer concentrations in endmembers and analytical errors, standard deviation was used if multiple samples were available and the analytical errors of 1% for ionic composition and 0.2‰ for δ 18 O were used for a single sample.…”