Coupled uptake dynamics of dissolved inorganic nitrogen and soluble reactive phosphorus in a moderately nutrient-polluted stream

Abstract

Although research on nutrient uptake in small streams has been extensive over the past 20 years, much less is known about coupled nutrient uptake, especially in a moderately nutrient-polluted stream. So, we selected a moderately nutrient-polluted stream (Zhangwa Creek) in Chaohu Lake basin, China, to examine the dynamics of coupled dissolved inorganic nitrogen (DIN, herein, ammonium nitrogen (NH₄-N) plus nitrate nitrogen (NO₃-N)) and soluble reactive phosphorus (SRP) uptake by performing a range of single- and dual-nutrient instantaneous additions experiments. The whole-reach uptake rate and saturation kinetics of nitrogen (N) and phosphorus (P) individually were estimated following the tracer additions for spiraling curve characterization (TASCC) model, and the dynamics of coupled DIN and SRP uptake in dual-nutrient additions were characterized through the response surface model. Comparisons of the ambient uptake length in single additions between DIN and SRP consistently indicated a slight P limitation. In contrast, results from response surfaces showed that the stream was likely limited by N or P or co-limited by them. Our results revealed that DIN and SRP uptake coupling effects could occur in a moderately nutrient-polluted stream. Moreover, we found that water temperature and pH might be central environmental variables influencing DIN and SRP uptake in the stream.

Graphical abstract

Keywords

coupled uptake dynamics / nutrient uptake / TASCC / moderately nutrient-polluted stream

1 Introduction

Eutrophication of inland and coastal waters caused by excess nitrogen (N) and phosphorus (P) poses an enormous threat to aquatic ecosystems (Conley et al., 2009; Schindler et al., 2016; Le Moal et al., 2019). Accounting for a large proportion of the total length of the fluvial system (Horton, 1945), headwater streams flowing through agricultural regions carry large amounts of N and P, which profoundly influences downstream water quality (Alexander et al., 2007). However, numerous studies have demonstrated that headwater streams could considerably reduce the export of N and P loads in stream flow to downstream ecosystems and effectively regulate downstream water quality via nutrient uptake (Royer et al., 2004; Bernal et al., 2012; Finkler et al., 2018).

The response of in-stream nutrient uptake to short-term nutrient additions should be closely related to a stream’s proximity to uptake saturation (i.e., nutrient availability over biotic demand) (Earl et al., 2006). The biotic uptake efficiency generally decreases with increasing N and P inputs (Mulholland et al., 2008; Li et al., 2020) when in-stream N and P concentrations approximate the saturation level, such as in chronically nutrient-rich or heavily nutrient-polluted streams. Accordingly, these streams approaching saturation can hardly take up additional nutrients (Li et al., 2020). Recently, much attention has been paid to assessing the potential of nutrient uptake in nutrient-poor or pristine streams (Covino et al., 2010a; Griffiths and Hill, 2014; Gibson et al., 2015; Piper et al., 2017) since these streams are well below saturation. Although these streams have great potential for nutrient uptake, they receive far lower N and P loads from the surrounding areas than those of nutrient-polluted streams (e.g., mesotrophic or eutrophic streams). By contrast, moderately nutrient-polluted streams (i.e., mesotrophic streams) generally receive considerable N and P and have immense uptake potential of nutrients (Weigelhofer et al., 2018), such as streams in an agricultural area or near the downstream of the area. In recent years, the role of moderately nutrient-polluted streams in reducing downstream nutrient pollution has become a focus for water environment managers and policymakers (Li et al., 2014; Weigelhofer et al., 2018; Namaalwa et al., 2020). However, there remains considerable uncertainty in the mechanism of nutrient uptake in moderately nutrient-polluted streams, such as interactions between N and P uptake and their influencing factors.

In stream ecosystems, nutrient limitation is vital to nutrient uptake (Tromboni et al., 2018). The nutrient will be considered limiting in a stream when its increasing concentration triggers an increase in biological activity (e.g., biotic uptake and biological assimilation) (Moore et al., 2013). Research has revealed that the uptake of non-limiting nutrients is closely tied to the availability of the limiting element (Schade et al., 2011). Specifically, increasing the concentration of the limiting essential nutrient will enhance its uptake and promote the uptake of non-limiting nutrients. In response, Piper et al. (2017) found the coupling effects of nutrient uptake at an ecosystem level in nutrient-poor streams. Although previous studies have quantified coupled nutrient uptake employing added N and P (Gibson and O’Reilly, 2012; Griffiths and Johnson, 2018; Tromboni et al., 2018; Li et al., 2021b), few of them have focused on the added nutrient forms, which thereby cannot offer accurate information about in-stream N and P uptake. Bio-available nutrients generally have multiple components or forms, for example, N, in the water column, mainly including ammonium nitrogen (NH₄-N) and nitrate nitrogen (NO₃-N). The NO₃-N and NH₄-N, two primary dissolved inorganic nitrogen (DIN) species in streams, can be assimilated directly by organisms (Farías et al., 2020). Further, there may be preferential uptake of NH₄-N over NO₃-N in a stream due to NH₄-N uptake having relatively low energy consumption (Gibson et al., 2015). Therefore, considering NH₄-N only or NO₃-N only rather than both in the nutrient-addition experiments is likely to overlook the influences of ambient NH₄-N (or NO₃-N) on the uptake of added NO₃-N (or NH₄-N) in streams, especially for streams with high background con-centrations of NO₃-N and NH₄-N. As for P in the water column, soluble reactive phosphorus (SRP) is the predominant form, assimilated more readily and utilized efficiently by most aquatic organisms than other P-forms (Duan et al., 2011). Thus, considering NH₄-N and NO₃-N as added N and SRP as added P in the nutrient addition experiments to study the interaction between N and P uptake will help us better understand the mechanism of multiple nutrient uptake in streams.

The Michaelis-Menten (M-M) kinetic model was initially proposed to simulate the enzyme kinetics, but it has proved to be robust in describing dynamics of single-nutrient uptake in aquatic ecosystems such as streams and rivers (e.g., Earl et al., 2006; Demars, 2008; O’Brien and Dodds, 2010; Covino et al., 2012). However, regarding interactions between multiple nutrient uptakes, the M-M model is restricted because it cannot reflect how variability in nutrient concentrations and ratios affect uptake dynamics. Piper et al. (2017) first applied a three-dimensional response surface model (i.e., the extended Monod equation) to describe the total areal uptake rate of either N or P as a function of both of their concentrations. This model can describe the dual-nutrient uptake coupling process and quantitatively reflect co-limitation and multi-element demand patterns. Based on the surface model, Tromboni et al. (2018) constructed hypothetical response surfaces to illustrate how results of single- and dual-nutrient additions should behave when stream ecosystems are limited by N and P and co-limited by both nutrients. Nevertheless, since the response surface model is empirical, more research is needed to test its applicability in streams, especially in moderately nutrient-polluted streams.

For this purpose, we performed a range of whole-reach nutrient addition experiments in two contrasting seasons (spring and autumn) at Zhangwa Creek, a moderately nutrient-polluted headwater agricultural stream in the Nanfei River watershed in Chaohu Lake basin, China, to examine the coupled N and P uptake dynamics. Specifically, the goals of this study were to (i) analyze the nutrient saturation state and assess nutrient limitation in the studied stream; (ii) examine the coupled uptake dynamics between DIN (herein, NH₄-N + NO₃-N) and SRP, and analyze the influence of a particular nutrient addition on the uptake rate of another nutrient; and (iii) tentatively identify the significant environmental factors influencing coupled DIN and SRP uptake at the stream ecosystem level. To address these goals, we (i) conducted five sets of single- and dual-nutrient instantaneous addition experiments during two biologically contrasting seasons (spring and autumn); (ii) estimated ambient nutrient spiraling metrics and saturation kinetic parameters using the tracer additions for spiraling curve characterization (TASCC) approach (Covino et al., 2010b); (iii) employed the response surface model to illustrate how the results of adding nutrients alone and in combination in a moderately nutrient-polluted stream would behave; and (iv) analyzed the relationship between nutrient uptake parameters in dual-nutrient additions and main environmental factors. We expected that the outcomes of this work would expand our understanding of interactions between DIN and SRP cycling in a moderately nutrient-polluted stream.

2 Study area

Nanfei River is located in the north-west of the Chaohu Lake basin. It flows through Hefei City from north-west to south-east and finally into the lake. The river has a total length of approximately 70 km and a drainage area of about 1700 km². This region’s annual mean temperature is about 15.7°C, and annual mean precipitation is approximately 1000 mm.

Zhangwa Creek is a first-order agricultural headwater stream of the Nanfei River, situated on the northern edge of the urban area of Hefei City (Fig. 1). The creek covers a 5.0 km² catchment, with approximately 2.3 km in length and 0.73–1.49 m in water width. As an agricultural headwater stream, Zhangwa Creek is chronically exposed to high N and P loads from the upstream agricultural region. There is a reservoir (Taochonghu Reservoir) about 1 km upstream of the creek, with a storage capacity of about 6.51 × 10⁵ m³. The reservoir’s outflow flows into a pond, and then feeds the creek. Two roughly parallel elevated railways are adjacent to the creek, and the catchment consists mainly of abandoned farmland. The experimental stream reach is composed of several riffle-pool sections. Additionally, benthic sediments mainly comprise rock block and gravel substrates, with minor contributions of clay, silt, leaf litter, and other debris. During growing seasons, massive herbaceous vegetation covers the stream banks, while only a few Brasenia schreberi J.F.Gmel. and Alternanthera philoxeroides (Mart.) Griseb. are scattered in the stream channel.

A representative 140-m long stream reach was selected in the middle of Zhangwa Creek in the present study, which was crossed over by two elevated railways at the upstream and downstream ends of the reach, respectively. The stream section was 0.85 to 1.42 m in water surface width, 0.20 to 0.60 m in water depth, and about 0.9 to 1.4 m of incision. The flow velocity ranged from 0.13 to 0.42 m·s⁻¹, and discharge varied between 0.078 and 0.102 m³·s⁻¹. Ambient physical and chemical characteristics of Zhangwa Creek are presented in Table 1.

3 Materials and methods

3.1 Tracer experimental design

We conducted five sets of instantaneous tracer addition experiments in the studied reach from 22 October, 2018 to 6 May, 2019 under base-flow conditions. Potassium bromide (KBr) and sodium chloride (NaCl) were selected as conservative tracers. Potassium nitrate (KNO₃) and ammonium chloride (NH₄Cl) were added as N and potassium phosphate (KH₂PO₄) was added as P. Each set of tracer experiments was performed following the tracer additions for spiraling curve characterization (TASCC) method (Covino et al., 2010b) and included single- and dual-nutrient additions. The single nutrient additions comprised single N additions (i.e., KNO₃ and NH₄Cl combined with KBr) and single P additions (i.e., KH₂PO₄ combined with NaCl). The dual nutrient additions were conducted as a combination of the first two but added with a time lag (Piper et al., 2017; Li et al., 2021b). In all experiments, the order of nutrient additions was single N addition, single P addition, and dual nutrient additions. In all dual additions, N was added first. The time interval (i.e., the time lag of 3–10 min in the studied reach) between the addition of N and P in the dual additions was set as half of the time when the conductivity of the water sample reached the peak level in the corresponding single-nutrient addition test to expand the range of the concentration ratio of N:P (Piper et al., 2017). The resting time between two additions was at least one hour to reduce the effect of the prior tracer residues as far as possible. In addition, KNO₃ and NH₄Cl were added with a fixed NO₃-N/NH₄-N molar ratio of 1:1 to alleviate the influence of a relative shortage of either NO₃-N or NH₄-N on the interactions between NO₃-N (or NH₄-N) and SRP. For better comparison, an equivalent amount of nutrients and conservative tracers were added in the same set of single- and dual-nutrient addition tests. One day before tracer experiments, reach characteristics (e.g., water quality and hydrologic conditions) were investigated, and subsequently, the tracer dosages were estimated according to the background nutrient concentrations and stream flow.

We first dissolved the nutrient and conservative tracer into roughly 10 L of stream water in each addition using a plastic bucket. Then, we instantaneously poured the resulting injection solution into the stream at the upstream head of the studied reach. To ensure thorough mixing, we positioned the injection site in the middle of the narrow channel, where the flow velocity was relatively high. Finally, we established the sampling site at the downstream end of the reach. We monitored the conductivity in real-time about 5 m upstream of the sampling site and grabbed each water sample per minute when the conductivity began to rise. We collected 30 to 45 samples per release experiment, and the specific number of samples depended on the flow velocity.

Water samples were stored at 4°C for transport to the laboratory, filtered with 0.45 μm glass fiber filter membrane, and analyzed within 24 h. Br⁻ and chloride (Cl⁻) concentrations were measured using ion-selective electrodes. The concentration of NO₃-N was measured by ultraviolet spectrophotometry. The concentration of NH₄-N was measured using Nessler’s reagent colorimetry. The concentration of SRP was analyzed using potassium persulfate oxidation spectrophotometry. Three replicate samples were set in each sample test process. All streamwater samples’ analysis and testing methods refer to the State Environmental Protection Administration of China (SEPAC) (2002).

On the day of all nutrient-addition experiments, three replicate water samples were collected to characterize background concentrations of Br⁻, Cl⁻, SRP, NO₃-N, and NH₄-N before the start of the experiments. Water temperature (T), pH, electrical conductivity (EC) and total dissolved solids (TDS) were determined in situ using a pH/EC/TDS Tester. The flow velocity and discharge were measured in situ using a portable flow meter in several sections of the studied stream reach. The length, water depth, width, and incision were measured in situ by tapes in several sections of the studied stream reach.

3.2 Models and methods

The TASCC model proposed by Covino et al. (2010b) has been successfully utilized to characterize the dynamic process of nutrients uptake in small streams (Covino et al., 2012; Arce et al., 2014; Bracken et al., 2015; Piper et al., 2017; Finkler et al., 2021). Here, the longitudinal uptake rate of added nutrient X (k_w-add-X, m⁻¹) (DIN or SRP, represented by X; and Br⁻ or Cl⁻, represented by R) was calculated by Eq. (1), and the negative inverse of k_w-add-X was the uptake length of added nutrient X (S_w-add-X) for each time point:

(1)k_{\text{w-add-}X}={\displaystyle \frac{\ln {\displaystyle \frac{C_{\text{bc-}X}}{C_{\text{bc-}R}}}-\ln {\displaystyle \frac{C_{\text{add-}X}}{C_{\text{add-}R}}}}{L}},

where ln is the natural logarithm, C_bc-X/C_bc-R is the concentration ratio of nutrient X to conservative tracer ion R in each grab sample after the background corrected, dimensionless; C_add-X/C_add-R is the concentration ratio of nutrient X to conservative tracer ion R in the mixed solution, dimensionless; and L is the distance between the dosing point and the sampling point, m.

The ambient uptake length of added nutrient X (S_w-amb-X, m) was calculated by regressing all S_w-add-X throughout the breakthrough curve (BTC) against the total nutrient concentration (C_tot-X, mg·L⁻¹) and back-extrapolating to the ambient concentration (C_amb-X, mg·L⁻¹). Then the ambient uptake rate of added nutrient X (U_amb-X, g·m⁻²·s⁻¹) was calculated by Eq. (2):

(2)U_{\text{amb-}X}=\frac{Q\times C_{\text{amb-}X}}{w\times S_{\text{w-amb-}X}},

where U_amb-X is the ambient areal uptake rate of added nutrient X, g·m⁻²·s⁻¹; Q is stream discharge, m³·s⁻¹; C_amb-X is the ambient concentration of added nutrient X, mg·L⁻¹; S_w-amb-X is the ambient uptake length of added nutrient X, m; and w is the average wetted stream width for experimental sections, m.

The areal uptake rate of added nutrient X (U_add-X, g·m⁻²·s⁻¹) for each grab sample was calculated by Eq. (3):

(3)U_{\text{add-}X}=\frac{L\times d\times C_{\text{add-}X}}{t\times S_{\text{w-add-}X}},

where t is the travel time since the slug addition and a specific sampling time on the BTC, s; d is the average stream depth, m; and C_add-X is the geometric mean value of background-corrected and ‘conservative’ concentrations of added nutrient X in a grab sample, mg·L⁻¹. The ‘conservative’ concentration of added nutrient X is the concentration expected had the nutrient traveled conservatively (i.e., no uptake, the maximum that could arrive at a sampling site), which was calculated as the product of C_bc-R and C_add-X/C_add-R. The total areal uptake rate of added nutrient X (U_tot-X) for each grab sample equals the sum of ambient and added uptake rate of nutrient X.

In this study, the M-M model was applied for fitting the uptake kinetics of DIN and SRP (Dugdale, 1967), and it was expressed as follows:

(4)U_{\text{tot-}X}=U_{\text{max-}X}\times \frac{C_X}{K_{\text{m-}X}+C_X},

where U_tot-X is the total areal uptake rate of nutrient X, g·m⁻²·s⁻¹; U_max-X is the maximum uptake rate of nutrient X, g·m⁻²·s⁻¹; C_X is the concentration of nutrient X, mg·L⁻¹; and K_m-X is the half-saturation constant of nutrient X, mg·L⁻¹.

For comparison with results from the single nutrient additions, herein the M-M model was also used to characterize the uptake kinetics of DIN and SRP, respectively, in the dual-nutrient additions, without considering mutual interaction between DIN and SRP. Furthermore, the response surface model (Piper et al., 2017) was used to quantitatively describe the characteristics of the coupled uptake kinetics between DIN and SRP. The equation is expressed as follows:

(5)U_{\text{tot-}X}=U_{\text{max-}X}\times \frac{C_{\text{tot-N}}}{K_{\text{m-N}}+C_{\text{tot-N}}}\times \frac{C_{\text{tot-P}}}{K_{\text{m-P}}+C_{\text{tot-P}}},

where C_tot-N and C_tot-P are the concentrations of DIN and SRP, respectively, mg·L⁻¹; and K_m-N and K_m-P are the half-saturation constants of DIN and SRP, respectively, mg·L⁻¹.

3.3 Data processing and analysis

The Origin 2021 (OriginLab Corporation, USA) software was used to plot basic graphs, perform the Shapiro-Wilk (S-W) test, and conduct correlation and regression analyses. Before conducting correlation analysis, the Shapiro-Wilk (S-W) test was used to check the normality of the data. The MatLab R2018b (MathWorks, Natick, MA, USA) software was used to draw the coupled uptake kinetic surface (i.e., response surfaces). The response surfaces are constructed based on data from single- and dual-nutrient additions, including concentrations and total areal uptake rates of SRP and DIN. The data from single nutrient additions build boundaries for N or P surfaces. More details on constructing response surfaces are described in Piper et al. (2017) and Tromboni et al. (2018). All model fittings, including the traditional M-M and response surface models, were constructed using the nonlinear least square method. This method uses the Trust-Region algorithm with a Bisquare weights scheme to adjust the coefficients. It allows us to solve complex nonlinear problems more efficiently than the other methods and minimize the effect of outliers (Tiwari and Shanmugam, 2013).

A regulatory coefficient (H) was used to assess the degree of homeostasis of whole-reach uptake N:P (U_N:U_P) in response to changes in molar N:P supply (Schade et al., 2011). This theory extends the organismal homeostasis theory (Sterner and Elser, 2002). H values were calculated as the inverse slope of a log-log plot of U_N:U_P versus molar N:P supply.

To identify the main environmental factors affecting N and P coupling uptake, we first conducted a Spearman’s correlation analysis between coupled uptake parameters (i.e., ambient nutrient spiraling metrics obtained in the dual-nutrient additions and nutrient uptake kinetic parameters of response surface models) and the observed environmental factors (i.e., Q, T, pH, and ambient DIN and SRP concentrations) of the studied stream reach in the five tracer tests, based on the non-normal distribution data. Then, we further conducted regression analysis on the variables with significant correlation.

4 Results

4.1 Comparisons of M-M kinetic parameters between single and dual nutrient additions

For both single and dual nutrient additions, the pattern of U_tot-X increasing with C_tot-X well conformed to the M-M model (R² > 0.9; Table 2). In the five tracer experiments, U_tot-X values of dual-nutrient additions were consistently higher than those of corresponding single-nutrient additions (Fig. 2). Regardless of DIN or SRP, U_max-X was smaller in single additions than in dual additions, while K_m-X exhibited an opposite pattern (Table 2).

4.2 Comparisons of ambient nutrient spiraling metrics between single and dual nutrient additions

Across all the tracer experiment dates, S_w-amb-X was consistently shorter in single additions than in dual additions, regardless of DIN or SRP (Table 3). Compared with S_w-amb-X in single additions, S_w-amb-DIN decreased by 13.45%–45.04%, and S_w-amb-SRP declined by 14.23%–40.89% in dual additions (Table 3). S_w-amb-DIN was consistently longer than S_w-amb-SRP in single nutrient additions (Table 3). In addition, U_amb-DIN was invariably greater in dual additions than in single additions, while U_amb-SRP showed a similar pattern in most cases (Table 3).

4.3 Model fitting of coupled DIN and SRP uptake kinetics

The response surfaces showed that the uptake of either nutrient responded differently to the presence of the other (Fig. 3). Specifically, P surfaces indicated that SRP uptake rates were most responsive to DIN additions on 11/02/2018 and 03/12/2019 (Figs. 3(d) and 3(h)) while less responsive on the other dates, regardless of SRP concentrations (Figs. 3(b), 3(f), and 3(j)). In contrast, N surfaces showed two opposite effects of SRP additions on DIN uptake rates, namely, stimulation and inhibition effects. For example, DIN uptake significantly decreased as SRP concentrations elevated from the background on 11/02/2018 (Fig. 3(c)). It exhibited an opposite pattern on 11/29/2018 (Fig. 3(e)). K_m-SRP was lower than K_m-DIN in all N surfaces (Table 4). Conversely, K_m-DIN was consistently lower than K_m-SRP in P surfaces except on 11/02/2018 (Table 4).

4.4 Impact of relative nutrient supply on N and P uptake in dual nutrient additions

There was a significantly positive linear relationship between ln(U_N:U_P) and ln(N:P of supply) (Fig. 4). The variation range of molar N:P ratios on 10/22/2018 and 11/02/2018 was greater than those on 11/29/2018, 03/12/2019 and 05/06/2019 (Fig. 5). Specifically, molar N:P ratios on 10/22/2018 and 11/02/2018 ranged from 2 to 228 and 5 to 270, respectively, while those on 11/29/2018, 03/12/2019 and 05/06/2019 were from 6 to 85, 6 to 116, and 7 to 57, respectively. All slopes of linear regression curves of ln(U_N:U_P) versus ln(N:P of supply) were less than 1 (Fig. 4). Except for 05/06/2019 (slope = 0.35), the slopes were similar on the other dates (0.65 to 0.76; Fig. 4). H values suggested that the whole-reach U_N:U_P in response to changes in molar N:P supply was most homeostatic on date 05/06/2019 (H = 2.86).

4.5 Relationship between coupled uptake indicators and environmental variables

We observed a significantly negative correlation between S_w-amb-SRP in dual nutrient additions and water temperature (Spearman; P < 0.05) and a significantly positive correlation between U_max-DIN in dual nutrient additions and water pH (Spearman; P < 0.05). Except for these results, no significant relations were found between other variables (Fig. 6). Further, the regression analysis revealed their potential relationship: S_w-amb-SRP in dual nutrient additions shortened linearly with the elevation of water temperature (Fig. 7(a)) and U_max-DIN in dual nutrient additions became nonlinearly (i.e., power function) larger with higher water pH (Fig. 7(b)).

5 Discussion

5.1 Coupling of N and P uptake and nutrient limitation in the moderately nutrient-polluted stream

We observed higher U_max-X and lower K_m-X in dual additions compared to single additions in all tracer experiments conducted at Zhangwa Creek based on the fitting results of the M-M model (Table 2). Generally, a higher U_max-X value indicates that the ecosystem has a greater capacity to take up nutrient X. A lower K_m-X value means that ecosystems have a higher affinity to it. Therefore, when N and P were added simultaneously in the stream, the ecosystem’s uptake capacity and affinity to either nutrient increased compared to when they were added individually. In other words, the stream exhibited a significant interaction between N and P uptake.

The limiting nutrient should always have a shorter ambient uptake length than the non-limiting nutrient (Cross et al., 2005; Tromboni et al., 2018). In our trials, S_w-amb-DIN was longer than S_w-amb-SRP in single nutrient additions, suggesting a probably P-limited system. Nevertheless, whether there is an N limitation according to the comparison remains unknown.

A “curved” phenomenon on the upper right edge of N surfaces and the upper left edge of P surfaces could be used to describe visually and identify rapidly the pattern that either nutrient uptake was strongly influenced by the presence of the other nutrient at relatively low concentrations (i.e., nutrient limitation; Tromboni et al., 2018). The higher the degree of edge curvature, the more substantial the nutrient limitation. It is evident that all response surfaces displayed varying degrees of curved edges (Fig. 3). However, this method of identifying the nutrient limitation is not rigorous, as the surface in the fitting process may be extended, resulting in the original data being unable to support the bending result. Moreover, we found a “curved” phenomenon (e.g., Fig. 3(c)), which is contrary to our expectations and prior studies (Piper et al., 2017; Tromboni et al., 2018; Li et al., 2021b). Therefore, accurately assessing the nutrient limitation requires comprehensively considering response surfaces and fitting parameters. Based on the fitting parameters of response surfaces, we found that except for the P surface on 11/02/2018, the confidence intervals of the half-saturation constant for either SRP or DIN in opposite nutrient surfaces overlap with zero. This result indicated that the stream system was likely to be limited by N or P or co-limited by N and P.

Compared to oligotrophic streams, mesotrophic streams (i.e., moderately nutrient-polluted streams) typically exhibit U_max values for N and P that are one to two orders of magnitude higher, regardless of the nutrient combination method used (Table 5). This result may be related to the stream system’s biomass. Mesotrophic streams typically have higher biomass and, therefore, greater potential for nutrient uptake than oligotrophic streams (Weigelhofer et al., 2018). In addition, when nutrient combinations are considered, U_max-SRP is generally larger when adding NH₄-N and NO₃-N as N than when adding NH₄-N as N in mesotrophic streams (Table 5). As a result, only considering the interaction between single forms of nutrients may underestimate the nutrient uptake potential of streams. Contrary to our previous hypothesis, coupled N and P uptake paradigms were highly similar in oligotrophic and mesotrophic streams despite the varying combination of added nutrients (Piper et al., 2017; Li et al., 2021b). Specifically, N uptake surfaces consistently had a larger K_m for N than for P, whereas P uptake surfaces showed an opposite pattern (Table 5). These findings are likely linked to the relative availability of essential nutrients rather than absolute concentrations, as nutrient limitation is assessed by the stoichiometric ratio of elements (Tromboni et al., 2018). However, these findings and speculations still need further validation due to the lack of similar experimental data. Implementing the same nutrient-combination addition experiments in streams with different nutrient-polluted levels and expanding them to a larger time scale (e.g., months, seasons, years) will help reveal the similarities and differences in the N and P uptake coupling mechanisms among these streams.

It should be noted that the coupling of N and P uptake may be an outcome that integrates abiotic removal (e.g., abiotic absorption of P, precipitation of P, and volatilization of ammonia) (Reddy et al., 1999; Griffiths and Johnson, 2018) and biotic uptake (e.g., uptake and assimilation by benthic biota). However, the TASCC method we used here cannot strictly distinguish between biotic and abiotic contributions (Covino et al., 2010b; Piper et al., 2017). The coupling of N and P uptake during short-term tests (i.e., within one hour) appears not to be a result produced by a strong response of benthic biota in the stream ecosystem because the benthic biota is not likely to respond quickly to newly added nutrient levels over tens of minutes (Covino et al., 2010a; Piper et al., 2017; Li et al., 2021a). By contrast, temporary storage in transient storage zones is a potential mechanism elucidating N and P uptake coupling, as it commonly occurs in streams (Li et al., 2020; Li et al., 2021a). For example, the added N and P might be caught by the substances (e.g., biofilm, fine sediment particles, colloidal particles) on the riverbed surface (Battin et al., 2003; Cohen et al., 2013), namely the adsorption process, resulting in a decline in concentrations of N and P in the water column to some degree.

5.2 Relative availability of nutrients in stream ecosystems

Without external input, the background concentration of NO₃-N in aquatic environments is generally higher than that of NH₄-N, as the latter is more prone to oxidation (e.g., nitrification; Merbt et al., 2017). Consistent lower background concentration of NH₄-N than NO₃-N, preferential assimilation of NH₄-N over NO₃-N by aquatic organisms, and their different uptake pathways cause a particular competitive relationship in their in-stream uptake processes (Pastor et al., 2013; Gibson et al., 2015; Ribot et al., 2017). However, we noticed that the ambient concentration of NH₄-N and NO₃-N was high at Zhangwa Creek, and that ambient NH₄-N concentration was higher than NO₃-N in most cases (Table 1), possibly due to agricultural fertilization in the upstream watersheds. Notably, the additional NO₃-N generated by nitrification would mask the uptake of NO₃-N. Adding NH₄-N and NO₃-N with a fixed molar ratio of 1:1 in all tracer experiments is expected to keep the concentrations of NH₄-N and NO₃-N at relatively high levels throughout their flowing downstream, likely partially counter-balancing the adverse conditions of NO₃-N uptake. However, here, we only considered the response of the overall DIN uptake to SRP addition, as we could not accurately evaluate NO₃-N uptake.

Our tracer experiments combined DIN and SRP satisfactorily characterized the kinetics of coupled N and P uptake. However, there were still some areas for improvement in revealing the biotic mechanisms of nutrient uptake and distinguishing between abiotic removal and biotic uptake. Considering only one molar dosage ratio of NH₄-N/NO₃-N (i.e., 1:1) in a moderately nutrient-contaminated stream is insufficient to fully reflect the characteristics of DIN and SRP uptake dynamics varying with the stoichiometric ratios of N:P. Thus, various ratios of NH₄-N to NO₃-N should be considered in future work to better explore the relations between the N:P stoichiometry and nutrient uptake in streams.

In the dual-nutrient addition experiments, we found a significantly positive linear relationship between U_N:U_P and molar N:P supply (Fig. 4), indicating that short-term increases in nutrient concentrations in the water column could stimulate whole-reach nutrient uptake. Studies have led some to suggest that the uptake of nutrients in unsaturated streams should be stimulated with increasing inputs of nutrients (Covino et al., 2010a, 2010b; Griffiths and Johnson, 2018). Interestingly, the increasing rate between U_N:U_P and molar N:P was almost invariable, and U_N:U_P increased more slowly than supply molar N:P in all cases (i.e., slope < 1; Fig. 4). However, the slope of the relationship between U_N:U_P and molar N:P supply in the studied stream was lower than that of a research in forested streams (slope = 0.98, Tomczyk et al., 2023), probably because the latter has low background nutrient concentrations. Our results indicated that N and P uptake rates were elevated simultaneously by N increments relative to P. However, U_tot-SRP increased more quickly than U_tot-DIN, supporting a P-limited system. Because N concentrations are not constant in the comprehensive regression results of U_N:U_P vs. molar N:P, we cannot estimate the influence of variations of P concentrations on both nutrient uptake. Nonetheless, this defect does not affect our conclusion of P limitation in this stream.

5.3 Factors affecting DIN and SRP uptake coupling

Climatically, March, April, and May are the spring months, and October and November are the autumn months in Chaohu Lake basin, China. Our experiments showed a significantly negative correlation between S_w-amb-SRP and T at Zhangwa Creek over a relatively large set of time scales (Fig. 7(a)), indicating possibly a seasonal effect (e.g., stream metabolism) on in-stream P uptake (Hoellein et al., 2007). Besides, we also observed a significantly positive correlation between U_max-DIN and pH at Zhangwa Creek (Fig. 7(b)). This result indicated that higher pH facilitated DIN uptake in the stream. Surprisingly, the relationship between nutrient uptake parameters and nutrient concentration was weak, possibly related to weak nutrient limitation and insufficient data. Moreover, we did not find a significant relationship between Q and nutrient uptake parameters, which is inconsistent with previous studies (Tromboni et al., 2018; Weigelhofer et al., 2018). Discharge usually plays a vital role in nutrient uptake in small streams. According to the results from García et al. (2017), an increase in discharge might reduce the relative size of the transient storage zone, thereby decreasing the uptake of nutrients in streams. However, the actual factors affecting DIN and SRP uptake may exceed our findings mentioned above due to limitations in the number of experiments.

It is worth noting that experimental process and model selection may also impact observed results or conclusions. The added nutrients mainly undergo two processes, namely the water column and the hyporheic zone processes (Li et al., 2021a), characterized by the rising and falling limbs of the breakthrough curve. The rising limb reflects the water column process, and the falling limb reflects the water column and hyporheic zone processes. The time-lagged addition method in dual-nutrient addition experiments means that only the water column process of the second nutrient influences the first nutrient. Therefore, the order of adding DIN and SRP may produce different outcomes (Piper et al., 2017). In addition, the response surface model we have chosen can only describe the positive coupling effect between two nutrients. When there is inhibition between them, this model is not applicable (Sperfeld et al., 2016). In our trials, we found inhibitory effects of SRP on DIN uptake, which may be an artifact of the model fitting (Tromboni et al., 2018).

6 Conclusions

Our study highlights the importance of studying N and P uptake coupling in the moderately nutrient-polluted stream (i.e., mesotrophic stream) by considering DIN (i.e., NH₄-N + NO₃-N) as added N in the tracer experiments. We found a significant interaction between DIN and SRP uptake at Zhangwa Creek, although its background concentrations of N and P were high. By comparing with prior studies, we found a similar pattern of coupled N and P uptake in oligotrophic and mesotrophic streams without considering the combination of added nutrients. However, this finding still needs further validation by more nutrient addition experiments using the same combination of nutrients in different trophic streams.

References

[1] Alexander R B, Boyer E W, Smith R A, Schwarz G E, Moore R B (2007). The role of headwater streams in downstream water quality.J Am Water Resour Assoc, 43(1): 41–59

[2] Arce M I, von Schiller D, Gómez R (2014). Variation in nitrate uptake and denitrification rates across a salinity gradient in Mediterranean semiarid streams.Aquat Sci, 76(2): 295–311

[3] Battin T J, Kaplan L A, Denis Newbold J, Hansen C M E (2003). Contributions of microbial biofilms to ecosystem processes in stream mesocosms.Nature, 426(6965): 439–442

[4] Bernal S B, von Schiller D, Marti E, Sabater F (2012). In-stream net uptake regulates inorganic nitrogen export from catchments under base flow conditions.J Geophys Res, 117(G3): G00N05

[5] Bracken M E S, Hillebrand H, Borer E T, Seabloom E W, Cebrian J, Cleland E E, Elser J J, Gruner D S, Harpole W S, Ngai J T, Smith J E (2015). Signatures of nutrient limitation and co-limitation: responses of autotroph internal nutrient concentrations to nitrogen and phosphorus additions.Oikos, 124(2): 113–121

[6] Cohen M J, Kurz M J, Heffernan J B, Martin J B, Douglass R L, Foster C R, Thomas R G (2013). Diel phosphorus variation and the stoichiometry of ecosystem metabolism in a large spring-fed river.Ecol Monogr, 83(2): 155–176

[7] Conley D J, Paerl H W, Howarth R W, Boesch D F, Seitzinger S P, Havens K E, Lancelot C, Likens G E (2009). Controlling eutrophication: nitrogen and phosphorus.Science, 323(5917): 1014–1015

[8] Covino T, McGlynn B, Baker M (2010a). Separating physical and biological nutrient retention and quantifying uptake kinetics from ambient to saturation in successive mountain stream reaches.J Geophys Res, 115(G4): G04010

[9] Covino T P, McGlynn B L, McNamara R A (2010b). Tracer Additions for Spiraling Curve Characterization (TASCC): quantifying stream nutrient uptake kinetics from ambient to saturation.Limnol Oceanogr Methods, 8(9): 484–498

[10] Covino T, McGlynn B, McNamara R (2012). Land use/land cover and scale influences on in-stream nitrogen uptake kinetics.J Geophys Res, 117(G2): G02006

[11] Cross W F, Benstead J P, Frost P C, Thomas S A (2005). Ecological stoichiometry in freshwater benthic systems: recent progress and perspectives.Freshw Biol, 50(11): 1895–1912

[12] Demars B O L (2008). Whole-stream phosphorus cycling: testing methods to assess the effect of saturation of sorption capacity on nutrient uptake length measurements.Water Res, 42(10−11): 2507–2516

[13] Duan S W, Amon R, Bianchi T S, Santschi P H (2011). Temperature control on soluble reactive phosphorus in the Lower Mississippi River.Estuaries Coasts, 34(1): 78–89

[14] Dugdale R C (1967). Nutrient limitation in the sea: dynamics, identification, and significance.Limnol Oceanogr, 12(4): 685–695

[15] Earl S R, Valett H M, Webster J R (2006). Nitrogen saturation in stream ecosystems.Ecology, 87(12): 3140–3151

[16] Farías L, Faúndez J, Sanhueza-Guevara S (2020). Temporal dynamics of dissolved inorganic nitrogen (DIN) in the aphotic layer of a coastal upwelling system with variable dissolved oxygen.J Mar Syst, 209: 103087

[17] Finkler N R, Gücker B, Boëchat I G, Tromboni F, Thomas S A, Mendes L A, Lima D M F, Covino T, Emanuelson K, Ponce de León C, Fernandes Cunha D G (2021). Comparing spiraling- and transport-based approaches to estimate in-stream nutrient uptake length from pulse additions.Ecohydrology, 14(7): e2331

[18] Finkler N R, Tromboni F, Boëchat I G, Gücker B, Fernandes Cunha D G (2018). Nitrogen and phosphorus uptake dynamics in tropical Cerrado woodland streams.Water, 10(8): 1080

[19] García V J, Gantes P, Giménez L, Hegoburu C, Ferreiro N, Sabater F, Feijoó C (2017). High nutrient retention in chronically nutrient-rich lowland streams.Freshw Sci, 36(1): 26–40

[20] Gibson C A, O’Reilly C M, Conine A L, Lipshutz S M (2015). Nutrient uptake dynamics across a gradient of nutrient concentrations and ratios at the landscape scale.J Geophys Res Biogeosci, 120(2): 326–340

[21] Gibson C A, O’Reilly C M (2012). Organic matter stoichiometry influences nitrogen and phosphorus uptake in a headwater stream.Freshw Sci, 31(2): 395–407

[22] Griffiths N A, Hill W R (2014). Temporal variation in the importance of a dominant consumer to stream nutrient cycling.Ecosystems (N Y), 17(7): 1169–1185

[23] Griffiths N A, Johnson L T (2018). Influence of dual nitrogen and phosphorus additions on nutrient uptake and saturation kinetics in a forested headwater stream.Freshw Sci, 37(4): 810–825

[24] Hoellein T J, Tank J L, Rosi-Marshall E J, Entrekin S A, Lamberti G A (2007). Controls on spatial and temporal variation of nutrient uptake in three Michigan headwater streams.Limnol Oceanogr, 52(5): 1964–1977

[25] Horton R E (1945). Erosional development of streams and their drainage basins: a hydrophysical approach to quantitative morphology.Geol Soc Am Bull, 56(3): 275–370

[26] Le Moal M, Gascuel-Odoux C, Ménesguen A, Souchon Y, Étrillard C, Levain A, Moatar F, Pannard A, Souchu P, Lefebvre A, Pinay G (2019). Eutrophication: a new wine in an old bottle.Sci Total Environ, 651: 1–11

[27] Li A, Bernal S, Kohler B, Thomas S A, Martí E, Packman A I (2021a). Residence time in hyporheic bioactive layers explains nitrate uptake in streams.Water Resources Research, 57(2): e2020WR027646

[28] Li R Z, Fu Y, Xu D Q, Wang X H, Jin G Q (2021b). Assessing the potential and kinetics of coupled nutrients uptake in mesotrophic streams in Chaohu Lake basin, China.Environ Sci Pollut Res Int, 28(44): 62877–62890

[29] Li R Z, Yang J W, Qian J, Dong Y H, Tang W K (2014). Characteristics of nitrogen and phosphorus retention in two different channel forms in a typical headwater stream in the suburb of Hefei City, China.Environ Sci, 35(9): 3365–3372

[30] Li R Z, Wang Y Q, Qian J, Fu Y (2020). Evaluating transient storage and associated nutrient retention in a nutrient-rich headwater stream: a case study in Lake Chaohu Basin, China.Environ Sci Pollut Res Int, 27(6): 6066–6077

[31] Merbt S N, Bernal S, Proia L, Martí E, Casamayor E O (2017). Photoinhibition on natural ammonia oxidizers biofilm populations and implications for nitrogen uptake in stream biofilms.Limnol Oceanogr, 62(1): 364–375

[32] Moore C M, Mills M M, Arrigo K R, Berman-Frank I, Bopp L, Boyd P W, Galbraith E D, Geider R J, Guieu C, Jaccard S L, Jickells T D, La Roche J, Lenton T M, Mahowald N M, Marañón E, Marinov I, Moore J K, Nakatsuka T, Oschlies A, Saito M A, Thingstad T F, Tsuda A, Ulloa O (2013). Processes and patterns of oceanic nutrient limitation.Nat Geosci, 6(9): 701–710

[33] Mulholland P J, Helton A M, Poole G C, Hall R O, Hamilton S K, Peterson B J, Tank J L, Ashkenas L R, Cooper L W, Dahm C N, Dodds W K, Findlay S E G, Gregory S V, Grimm N B, Johnson S L, McDowell W H, Meyer J L, Valett H M, Webster J R, Arango C P, Beaulieu J J, Bernot M J, Burgin A J, Crenshaw C L, Johnson L T, Niederlehner B R, O’Brien J M, Potter J D, Sheibley R W, Sobota D J, Thomas S M (2008). Stream denitrification across biomes and its response to anthropogenic nitrate loading.Nature, 452(7184): 202–205

[34] Namaalwa S, van Dam A A, Gettel G M, Kaggwa R C, Zsuffa I, Irvine K (2020). The impact of wastewater discharge and agriculture on water quality and nutrient retention of Namatala Wetland, Eastern Uganda.Front Environ Sci, 8: 148

[35] O’Brien J M, Dodds W K (2010). Saturation of NO₃^- uptake in prairie streams as a function of acute and chronic N exposure.J N Am Benthol Soc, 29(2): 627–635

[36] Pastor A, Peipoch M, Cañas L, Chappuis E, Ribot M, Gacia E, Riera J L, Martí E, Sabater F (2013). Nitrogen stable isotopes in primary uptake compartments across streams differing in nutrient availability.Environ Sci Technol, 47(18): 10155–10162

[37] Piper L R, Cross W F, McGlynn B L (2017). Colimitation and the coupling of N and P uptake kinetics in oligotrophic mountain streams.Biogeochemistry, 132(1−2): 165–184

[38] Reddy K R, Kadlec R H, Flaig E, Gale P M (1999). Phosphorus retention in streams and wetlands: a review.Crit Rev Environ Sci Technol, 29(1): 83–146

[39] Ribot M, von Schiller D, Marti E (2017). Understanding pathways of dissimilatory and assimilatory dissolved inorganic nitrogen uptake in streams.Limnol Oceanogr, 62(3): 1166–1183

[40] Royer T V, Tank J L, David M B (2004). Transport and fate of nitrate in headwater agricultural streams in lllinois.J Environ Qual, 33(4): 1296–1304

[41] Schade J D, Macneill K, Thomas S A, Camille McNeely F, Welter J R, Hood J, Goodrich M, Power M, Finlay J C (2011). The stoichiometry of nitrogen and phosphorus spiralling in heterotrophic and autotrophic streams.Freshw Biol, 56(3): 424–436

[42] Schindler D W, Carpenter S R, Chapra S C, Hecky R E, Orihel D M (2016). Reducing phosphorus to curb lake eutrophication is a success.Environ Sci Technol, 50(17): 8923–8929

[43] Sperfeld E, Raubenheimer D, Wacker A (2016). Bridging factorial and gradient concepts of resource co-limitation: towards a general framework applied to consumers.Ecol Lett, 19(2): 201–215

[44] State Environmental Protection Administration of China (SEPAC) (2002). Water and Exhausted Water Monitoring Analysis Method (4th ed). Beijing: China Environment

[45] Sterner R W, Elser J J (2002). Ecological Stoichiometry. Princeton: Princeton University Press

[46] Tiwari S P, Shanmugam P (2013). An optical model for deriving the spectral particulate backscattering coefficients in oceanic waters.Ocean Sci, 9(6): 987–1001

[47] Tomczyk N J, Rosemond A D, Kominoski J S, Manning D W P, Benstead J P, Gulis V, Thomas S A, Hotchkiss E R, Helton A M (2023). Nitrogen and phosphorus uptake stoichiometry tracks supply ratio during 2-year whole-ecosystem nutrient additions.Ecosystems (N Y), 26(5): 1018–1032

[48] Tromboni F, Thomas S A, Gücker B, Neres-Lima V, Lourenço-Amorim C, Moulton T P, Silva-Junior E F, Feijó-Lima R, Boëchat I G, Zandonà E (2018). Nutrient limitation and the stoichiometry of nutrient uptake in a tropical rain forest stream.J Geophys Res Biogeosci, 123(7): 2154–2167

[49] Weigelhofer G, Ramiao J P, Puritscher A, Hein T (2018). How do chronic nutrient loading and the duration of nutrient pulses affect nutrient uptake in headwater streams.Biogeochemistry, 141(2): 249–263

RIGHTS & PERMISSIONS

Higher Education Press