Nitrate uptake in an agricultural stream estimated from high-frequency, in-situ sensors

Abstract

Real-time, continuous, in situ water quality sensors were deployed on a fourth-order Iowa (U.S.) stream draining an agricultural watershed to evaluate key in-stream processes affecting concentrations of nitrate during a 24-day late summer (Aug-Sep) period. Overall, nitrate-nitrogen (NO-N) concentrations declined 0.11 mg L km, or about 1.9% km and 35% in total across 18 km. We also calculated stream metabolic rates using in situ dissolved oxygen data and determined stream biotic N demand to be 108-117 mg m day. From this, we estimate that 11% of the NO-N concentration decline measured between two in-situ sensors separated by 2 km was a result of biotic NO-N demand, while groundwater NO-N data and estimates of groundwater flow contributions indicate that dilution was responsible for 53%. Because the concentration decline extends linearly across the entire 18 km of stream length, these processes seem consistent throughout the basin downstream of the most upstream sensor site. The nitrate-dissolved oxygen relationship between the two sites separated by 2 km, calculations of biotic NO-N demand, and diurnal variations in NO-N concentration all indicate that denitrification by anaerobes is removing less NO-N than that assimilated by aquatic organisms unable to fix nitrogen for their life processes, and thus the large majority of the NO-N entering this stream is not retained or removed, but rather transported downstream.

Monitoring the riverine pulse: Applying high‐frequency nitrate data to advance integrative understanding of biogeochemical and hydrological processes

“…If assimilation driven by photosynthetic N uptake is a principal driver of diel NO 3 − patterns, then minima should occur during daylight and maxima during darkness. Many studies show such a diel pattern with peaks during evening to early morning and minima during afternoon to early evening (Cohen, Heffernan, Albertin, & Martin, ; Heffernan & Cohen, ; C. S. Jones et al, ; Pellerin et al, ; Rode et al, ). However, peak diel NO 3 − closer to the middle of the day is often observed when photosynthetic demand should be high (Burns et al, ; Duncan et al, ; Halliday et al, ; Rusjan & Mikoš, ).…”

“…Such insights are readily derived from comprehensive storm‐by‐storm c – Q analyses. As sensor records extend over increasingly longer periods, event‐based and composite c – Q analysis approaches can be compared, thus eliminating the possibility that differences in interpretation may have resulted from low sampling frequency and inadequate representation of variation with discharge and season (Duncan, Welty, et al, ; C. S. Jones, Kim, Wilton, Schilling, & Davis, ).…”

“…One possible reflection of these processes, persistent diel variation in stream NO 3 − concentrations, is also well recognized and described in previous studies that pre‐date the widespread deployment of high‐frequency NO 3 − sensors (Manny & Wetzel, ; Mulholland, ; Triska, Kennedy, Avanzino, Zellweger, & Bencala, ). Recent investigations, however, have applied high‐frequency NO 3 − measurements to gain extensive and unprecedented detail about diel patterns (Bende‐Michl et al, ; Burns, Miller, Pellerin, & Capel, ), and more recently, holistic or two station approaches have been developed to estimate total in‐stream NO 3 − losses in river networks (Jarvie et al, ; C. S. Jones et al, ; Miller et al, ; Wollheim, Mulukutla, Cook, & Carey, ). Here, we summarize recent advances in understanding the role of in‐stream processes by analyses of high‐frequency NO 3 − measurements and highlight remaining challenges to be met by new analyses.…”

“…The magnitude of detectable diel NO 3 − variation in stream and rivers can be highly variable, ranging from <10% of mean daily concentration (Burns et al, ), to >50% of the peak value (Moraetis et al, ). Many investigators state that diel NO 3 − patterns are present through visual data examination without formally quantifying these patterns (Bende‐Michl et al, ; Duncan et al, ; C. S. Jones et al, ; Rode et al, ), which may be adequate for many purposes. However, several new approaches for quantifying diel NO 3 − variation have been proposed recently in sensor‐based studies, facilitating comparisons at individual sites as well as across multiple locations (Aubert & Breuer, ; Burns et al, ; Moraetis et al, ; Rusjan et al, ).…”

Downstream evolution of wastewater treatment plant nutrient signals using high‐temporal monitoring

“…Koenig, Shattuck, Snyder, Potter, and McDowell (2017) Questions still remain about the relationship between nutrients and in-stream processing in anthropogenic systems. High-temporal sensors have been used to evaluate N patterns in agricultural streams, with low concentrations observed in the afternoon and high concentrations in the early morning due to higher assimilation during photosynthesis than denitrification at night (Jones, Kim, Wilton, Schilling, & Davis, 2018). High frequency logging has shown that variations in anthropogenic inputs can overprint diel N patterns, which can make predictions of load from grab samples uncertain (Carey, Wollheim, Mulukutla, & Mineau, 2014;Duan, Powell, & Bianchi, 2014;Pellerin et al, 2009).…”