Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Oct 5;57(41):15499–15510. doi: 10.1021/acs.est.3c04765

Rethinking Aerobic Respiration in the Hyporheic Zone under Variation in Carbon and Nitrogen Stoichiometry

Anna B Tureţcaia †,*, Vanessa A Garayburu-Caruso ‡, Matthew H Kaufman ‡,§, Robert E Danczak ‡, James C Stegen ‡,△, Rosalie K Chu ∥, Jason G Toyoda ∥, M Bayani Cardenas †, Emily B Graham ‡,⊥,*
PMCID: PMC10586321  PMID: 37795960

Abstract

graphic file with name es3c04765_0006.jpg

Hyporheic zones (HZs)—zones of groundwater–surface water mixing—are hotspots for dissolved organic matter (DOM) and nutrient cycling that can disproportionately impact aquatic ecosystem functions. However, the mechanisms affecting DOM metabolism through space and time in HZs remain poorly understood. To resolve this gap, we investigate a recently proposed theory describing trade-offs between carbon (C) and nitrogen (N) limitations as a key regulator of HZ metabolism. We propose that throughout the extent of the HZ, a single process like aerobic respiration (AR) can be limited by both DOM thermodynamics and N content due to highly variable C/N ratios over short distances (centimeter scale). To investigate this theory, we used a large flume, continuous optode measurements of dissolved oxygen (DO), and spatially and temporally resolved molecular analysis of DOM. Carbon and N limitations were inferred from changes in the elemental stoichiometric ratio. We show sequential, depth-stratified relationships of DO with DOM thermodynamics and organic N that change across centimeter scales. In the shallow HZ with low C/N, DO was associated with the thermodynamics of DOM, while deeper in the HZ with higher C/N, DO was associated with inferred biochemical reactions involving organic N. Collectively, our results suggest that there are multiple competing processes that limit AR in the HZ. Resolving this spatiotemporal variation could improve predictions from mechanistic models, either via more highly resolved grid cells or by representing AR colimitation by DOM thermodynamics and organic N.

Keywords: river corridor, flume, nutrient cycling, dissolved organic matter, FTICR-MS

Short abstract

Understanding factors that regulate dissolved organic matter metabolism through space and time in hyporheic zones will enhance our ability to predict changes in river corridor water quality and greenhouse gas emissions.

Introduction

Dissolved organic matter (DOM) is a major driver of biochemical reactions in riverine ecosystems and a central component of global carbon (C) and nutrient cycles.1−3 Estimates suggest that rivers receive around 5.1 Pg-C year–1 from terrestrial ecosystems,4 including significant inputs of DOM as leachate from leaf litter supplied through the river channel.5−7 As DOM moves through river corridors, it is gradually degraded by microorganisms to release greenhouse gases and to provide energy and nutrients necessary for metabolic processes. On a global scale, rivers are responsible for emitting approximately 1.8 Pg-C year–1 CO2 into the atmosphere,8 and hyporheic zones (HZs) can disproportionately contribute to DOM turnover and CO2 emissions.9−12 These subsurface domains of groundwater–surface water interactions are hotspots for biochemical reactions in rivers that may contribute up to 96% of river corridor respiration.13

Due to the dynamic nature of HZs, understanding the factors that regulate HZ DOM metabolism through space and time is a key gap in our ability to predict changes in river corridor water quality and greenhouse gas emissions. DOM metabolism has been previously linked to physicochemical controls including DOM quality and quantity, temperature, and microbiology.14−18 New molecular techniques, such as Fourier transform ion cyclotron mass spectrometry (FTICR-MS), allow unprecedented insight into the chemistry of DOM19,20 and its metabolism in rivers.21−23 For example, recent field studies suggest that the thermodynamic (or energetic) favorability of DOM, as well as its nitrogen (N) content, could be central to degradation of DOM in HZs.22−24 Graham et al.24 suggested that microbial communities may preferentially target both thermodynamically favorable DOM and organic N for metabolic activities, while Stegen et al.23 proposed that hydrologic mixing in the HZ results in degradation of the thermodynamically unfavorable DOM through a priming mechanism supported by organic N.

Uncertainties in the spatiotemporal evolution and the impact of nutrient availability, particularly N, on DOM decomposition inhibit the ability to accurately predict changes in the global C cycle.25,26 Therefore, there is an increasing need for understanding the effects of N availability on the HZ’s function as a biogeochemical reactor. Studies have shown the complex and heterogeneous nature of river corridors and the constituents they transport. For example, distributions of dissolved oxygen (DO) and N change with residence time in the HZ,27 C/N ratios gradually decrease from headwaters to estuaries,28 and inferred biochemical transformations of DOM diverge with nutrient-based selective processes.21 All of the above works highlight the importance of a realistic representation of the spatiotemporal dynamics of coupled DOM and N cycles in the river corridor.

The role of microbial nutrient limitations in regulating C turnover in natural environments remains an active area of research, largely because they are challenging to measure in situ.29 Considering that true measurements of nutrient limitations involve assessing the abilities of individuals to adjust to nutrient availability, relative changes in the ratio of C/N (i.e., elemental stoichiometry) are often used to infer the balance of C and N limitations.29−33 While we acknowledge that other elements such as phosphorus and sulfur may also limit biological activities, we focus on N because it is a predominant limiting nutrient in temperate inland rivers and because field observations of interactions between several limiting elements are exceedingly difficult without experimentally disrupting natural dynamics.34

This is evident from a recent small-scale, short-duration study by Garayburu-Caruso et al.,35 which proposed that integrated DOM and nutrient cycles are critical to regulating aerobic respiration (AR) in hyporheic environments through a trade-off between C and N limitations. The authors suggest that under C limitation, AR is regulated by DOM thermodynamics and that under N limitation, AR is regulated by organic N availability. While this conceptual model exposes the complex functioning of the HZs as hotspots for DOM cycling under varying nutrient limitations, previous work has not addressed the role of these controls across spatial and temporal fluctuations that are characteristic of the HZ. Therefore, it is unknown how multiple limitations interact to influence AR at the scale of the HZ. Answering this question is important when considering enhancement and expansion of process-based models that aggregate processes across centimeter scales to the scales of river corridors.

We used a large flume, ultrahigh-resolution analysis of DOM composition (FTICR-MS), dissolved C and N concentration measurements, and optical measurements of DO to investigate spatiotemporal relationships between changes in C and N availability, DOM pool evolution, and AR in the HZ. We assayed changes in channel and porewater composition across space and time, addressing porewater chemistry along and across flowpaths within a bedform. Because the HZ is a highly dynamic hydrologic and chemical environment, we propose that spatial gradients in the C/N ratio facilitate co-occurring dependencies of AR on DOM thermodynamics and N content over short distances in the HZ.

Methods

Experimental Setup

To mimic natural river conditions, including flowpaths in the HZ, while also controlling variability otherwise present in natural systems (temperature, DOM input, water chemistry, and flow rate), we used a large water recirculating flume (Figure S1) with natural river water (Lower Colorado River, Austin, Texas, USA) and clean play-sand (Pioneer Sands in Brady, Texas, USA). The size of the flume (5m (L) × 0.3m (W) × 0.7m (D)) allows HZ sampling at high spatial resolution while also mimicking natural stream dynamics. Hyporheic flow in and out of the sand was induced via manually shaped bedforms. For brevity, the stoss side of the bedform or the downwelling side of the HZ is referred to as “DW”, and the lee side of the bedform or the upwelling side of the HZ is referred to as “UW”. To maintain control over the availability of DOM, no particulate organic matter was added to the sand. Rigid insulating foam and a black plastic sheet covered the flume top and the walls to offset temperature and light variation, known to impact metabolic activity and DOM photodegradation.36 The system was given a five-month incubation period to populate the sand with microbes while periodically adding leaf leachate. After the incubation, for this experiment, we added 200 g of loosely packed, slightly crushed dry leaves in a permeable double bag into the flume channel for 24 h. This process allowed DOM to leach into the system while minimizing the input of particulates.

Sample Collection

Surface water and porewater samples were collected before the addition of leaves “hour 0”, at the time of removal of leaves “hour 24”, and at “hour 72”. More frequent sample collection was avoided to prevent porewater cross-contamination from surrounding flowpaths due to the relatively large porewater volumes necessary to complete analyses. To investigate relationships between the preferential utilization of DOM across HZ flowpaths and the spatiotemporal evolution of C/N ratios, sampling locations captured porewater composition in DW and UW spatial domains of the HZ between 10 and 60 cm depths at 10 cm intervals (Figure S1). Porewater samples were collected from a single bedform via the drive-point method using sterile syringes attached to stainless-steel mini piezometers with Tygon tubing equipped with a stopcock and sterile 0.22 μm polyethersulfone filters into precombusted VOA vials (Figure S2, preacidified for FTICR-MS and not acidified for nonpurgeable organic carbon (NPOC) and total nitrogen (TN) analysis). Sample collection was constrained to a single bedform to match the extent of the DO monitoring equipment. Channel water samples were collected directly using syringes. The pH was measured at the time of collection (range 8.24–7.69, Table S1). In total, we collected 39 samples between hour 0 and hour 72, as 3 channel water samples and 36 porewater samples from DW and UW combined.

Samples designated for FTICR-MS molecular analysis were shipped on blue ice to Pacific Northwest National Laboratory (Richland, Washington, USA) and frozen at −20 °C until analysis.

NPOC and TN Analysis

To evaluate gradients in the C/N ratio through space and time in the HZ, analysis for the NPOC fraction of total organic C (TOC) and TN was performed via combustion using an Apollo 9000 TOC/TN Analyzer (Teledyne Tekmar, Mason, Ohio, USA) equipped with an STS 8000 autosampler at The University of Texas, Austin (Austin, Texas, USA). Prior to combustion, samples were automatically acidified with 25% by volume H3PO4 and sparged with CO2-free air to remove inorganic C content, followed by combustion at 680 °C. Calibration standards for both TOC (1–50 ppm-C) and TN (1–10 ppm-N as nitrate) were prepared via dilution from standards purchased at RICCA Chemical Company (Arlington, Texas, USA). All samples were kept in the dark at 4 °C prior to the analysis.

Relative changes in the C/N ratio were used to assess shifts in C versus N limitations following previous studies.37−39 Stoichiometric ratios of elements have been used extensively as an alternative to experimental manipulations in order to estimate potential restrictions on microbial metabolic activities in field or fieldlike settings.29 This allows the observation of natural shifts in elemental limitations without changing ecosystem dynamics.

Fourier Transform Ion Cyclotron Mass Spectrometry

To assess patterns and changes in DOM composition in the flume, we used FTICR-MS. Water samples were thawed in the dark at 4 °C for 72 h. DOM from 15 mL of the acidified sample was extracted with PPL cartridges (Bond Elut), following Dittmar et al.40 Extracts were injected into a 12 T Bruker SolariX FTICR-MS instrument (Bruker, SolariX, Billerica, Massachusetts, USA) outfitted with a standard electrospray ionization source, located at the Environmental Molecular Sciences Laboratory (Richland, Washington, USA). Data were acquired in negative mode at +4.2 kV and an ion accumulation of 0.05 s from 100 to 900 m/z (mass to charge) at 4 M.

Briefly, 144 scans were coadded for each sample and peak picked with m/z values using BrukerDaltonik Data Analysis (version 4.2), as per Garayburu-Caruso et al.41 Formularity42 software was used to preprocess raw spectra, which included internal calibration and assignment of chemical formulas (S/N > 7, mass measurement error < 0.5 ppm) (Supporting Information, Figure S3). Further processing of FTICR-MS results was conducted using the R package ftmsRanalysis.43,44 Data processing included (1) removal of peaks outside of a high confidence m/z range (200–900 m/z) and/or with a 13C isotopic signature, (2) calculating molecular formula properties (e.g., Kendrick defect, double-bond equivalent, and modified aromaticity index), and (3) determining the chemical class to which a given metabolite belonged (e.g., van Krevelen classes predicted via O/C and H/C ratios).45 The resulting metabolite chemical classes were: “amino sugar-like”, “carbohydrate-like”, “condensed hydrocarbon-like”, “lignin-like”, “lipid-like”, “other-like”, “protein-like”, “tannin-like”, and “unsaturated hydrocarbon-like”. The suffix “-like” in chemical classes denotes the uncertainty of the van Krevelen classification method to distinguish between different molecules with the same H/C and O/C ratio.46 Peaks without a formula assignment were given a chemical class “None”.

Overall, FTICR-MS results were regarded as semiquantitative because of influences of the environment and instrument’s specifics on resulting peak intensity, as described in detail in a recent synthesis by Kew et al.47 Briefly, peak intensities are strongly influenced by variability in ionization efficiency among molecules, ion transmission and accumulation time, and interference between ions with similar frequencies in ion signal detection.42,47−51 Therefore, we converted intensities to a binary presence/absence format (i.e., 1 or 0) for all FTICR-MS data analysis.

Nonubiquitous DOM Molecules

To focus on temporal changes in DOM within each spatial domain (i.e., DW, UW, and the channel), we concentrated analysis on the subset of dynamic DOM molecules that did not occur across multiple time points (Supporting Information). Similar approaches are commonly applied to FTICR-MS data to reduce the impact of widely distributed molecules on observed changes through space, time, or experimental groups, largely due to the density of FTICR-MS data products. Briefly, we derived molecules, henceforth referred to as “nonubiquitous”, that did not repeat across temporal sampling events within a single spatial domain. For example, peaks present within DW at one sampling event, regardless of depth, but absent at all other sampling events in DW were considered nonubiquitous. These molecules were then grouped according to their chemical class (e.g., protein-like) and added. We underline that to infer temporal changes within domains, we used the presence of nonubiquitous molecules and not absolute or relative concentrations. For further information on the intersection of peaks across samples, see Figure S4.

DOM Thermodynamics

To infer the thermodynamic favorability of DOM pools from FTICR-MS data, we used the Gibbs free energy of the half-reaction of carbon oxidation (ΔGCox) as per LaRowe and Van Cappellen52 with corrections for temperature and pH = 7 proposed by Song et al.53 (Supporting Information). Laboratory temperatures fluctuated within a narrow range (17 and 20 °C). Therefore, ΔGCox was calculated using the depth-averaged HZ temperatures recorded at the time of sample collection (290.5 K at hour 0, 291.2 K at hour 24, and 292.5 K at hour 72; Table S2) using thermistors (Hobo U12 temperature logger, Onset, Massachusetts, USA). Throughout the experiment, DO concentrations in the flume were between 1.28 and 5.76 mg L–1, suggesting that oxygen was the terminal electron acceptor, which allows us to make direct comparisons across samples using ΔGCox. The ΔGCox calculations were not further adjusted for the pH measured at the time of sample collection (Table S1) and used pH = 7. The ΔGCox calculations excluded molecular peaks with a “none” chemical class assignment. Average Gibbs free energy (ΔGCox) was then calculated for each sample (distribution in Figure S5).

Inferred Biochemical Reactions

We inferred potential biochemical reactions from FTICR-MS spectra to provide information about processes possibly involved in regulating DOM metabolism. These inferred biochemical reactions were identified by mapping pairwise mass difference (m/z) between all observed peaks within a sample (Supporting Information, Figure S6) onto a data set of 1255 most frequently observed transformations (Table S3) following previous studies.21,24,41,54,55 Inferred biochemical reactions represent the gain or loss of specific molecules or the substitution of elements within a molecule (Supporting Information). For example, a mass difference of 1.031634 Da would putatively indicate a substitution H3O_1N1, which is associated with the gain/loss of ammonia, and a mass difference of 71.0371 Da would correspond to the gain/loss of the amino acid alanine. The calculation considered the interpeak mass difference within 1 ppm of the expected mass of a reaction to be a match. While we acknowledge that mass differences could arise spuriously, the narrow mass-difference margin for identifying the biochemical reactions is enabled by the ultrahigh resolution FTICR-MS data set and allows for their robust inference.56 One advantage of this approach is that it can capture biochemical reactions involving molecules that are otherwise hard to observe in environmental samples such as amino acids, which have high turnover times.

Furthermore, to evaluate the involvement of N-containing DOM in AR along and across HZ flowpaths, we filtered the set of inferred biochemical reactions to those containing N. We calculated the relative abundance of N-containing inferred biochemical reactions at each sampling depth in DW and UW and through time (Supporting Information, Table S4). This relative abundance of inferred N-containing biochemical reactions was then regressed against average DO concentrations across depths in DW and UW. The percent abundance of individual N-containing inferred biochemical reactions identified across all samples is reported in Table S5.

DO Monitoring

To infer AR, DO was monitored using a large planar optode—an in situ DO imaging system. Details regarding the construction of the planar optode, the specifics of the imaging system, and the conversions from ratiometric values obtained from red and green image channels to DO % saturation can be found in Kaufman et al.57 and Larsen et al.58 The optode calibration over a DO saturation range as close to 100 and 0% saturation was performed by using a FireSting O2 sensor with a temperature module (PyroScience, Aachen, Germany). Image processing and unit conversions were performed in Python 3 v3.8.559 using Pillow60 and NumPy61 libraries. Conversions to concentration incorporated depth-averaged temperature of the sand recorded with thermistors and by matching timestamps to the nearest 30 min between optode images and the temperature records (Table S2). This allowed some offset of temperature effects on DO concentration and on the planar optode which is temperature-sensitive.62

Further investigation of simultaneous association of microbial uptake of DO over time with both the thermodynamics and N content of DOM required DO data corresponding to locations for water sample collection. For this, we extracted ten 5 cm2 areas from the optode image, which captured DO concentrations corresponding to 10–50 cm sampling depths in DW and UW sampling domains. We obtained five points with average DO concentrations in DW and five points with average DO concentrations in UW for hours 0, 24, and 72 (DO distribution is shown in Figure S7). The 60 cm location was beyond the physical extent of the optode. Therefore, further correlations involving DO extend only to a 50 cm depth into the HZ. Average channel DO concentrations were calculated after masking portions of the image below the sediment–water interface.

Statistical Analysis

Data analysis and visualization were performed using R v4.163 and Python 3. Prior to analysis, FTICR-MS intensities were converted to presence/absence, as described above. Nonmetric multidimensional scaling (NMDS) for visualizing differences in DOM composition across the channel, DW, and UW spatial domains was performed using the R vegan package v2.5–664 with a Jaccard distance matrix constructed using vegdist and evaluated for multivariate differences with PERMANOVA using adonis. This evaluation for broad compositional differences of DOM across the spatial domains was based on all peaks detected by FTICR-MS regardless of formula assignment (i.e., including class “None”). The ordination plot was graphed using the ggplot2 package (v3.2.2).65

All other statistical analyses were performed in Python 3, using pandas,66 NumPy, sklearn,67 and SciPy68 libraries with graphical visualization via Matplotlib. This included: (1) linear regression analysis for identifying changes in NPOC and TN with depth; (2) linear regression analysis for ΔGCox and inferred N-containing biochemical reactions across DO concentrations; and (3) ANOVA followed by the Tukey HSD test for temporal differences in the average DO and ΔGCox values of samples from DW and UW (separately). Normality was assessed with the Shapiro–Wilk test. The threshold p-value for significance in all statistical tests was 0.05. The data package used in this study can be found at https://ess-dive.lbl.gov/ using identifier ess-dive-86ee624595adb17-20230925T224415467.

Results and Discussion

Developing predictive models that scale from local processes to represent river corridor metabolism requires careful consideration of spatiotemporal heterogeneity in physical, biological, or chemical controls. For example, microbial metabolic processes require a balance between the stoichiometric needs of microbial biomass and the stoichiometry of available resources in the environment. When environmental nutrient availability falls short of microbial nutrient demand, nutrient limitations can arise.69,70 While the exact thresholds that define nutrient limitations are challenging to quantify without assessing the abilities of individual microorganisms to adjust to nutrient availability,29,69 microbial demand for C and N is known to drive rates of organic matter decomposition.34,71,72 However, the current understanding of the relationships between C, N, DOM quantity and quality, and metabolism in rivers has mostly been derived from batch reactors, flow-through columns, or field-scale experiments and observations. Taking advantage of the elemental stoichiometry of C and N (C/N) as a proxy for microbial nutrient limitations,29,30,34 this work investigates the spatiotemporal relationships between aerobic metabolism, DOM, and the importance of C and N in the HZ. While we acknowledge that other elements may also constrain biological processes, we focused on N limitations due to recent work highlighting its possible interactions with DOM cycling in the river corridor.22−24,35 Collectively, this study provides an intermediate scale of evaluation to help link highly controlled small-scale laboratory experiments to natural river corridor dynamics.

Changes in DO, DOM Chemistry, and C/N Over Short Spatial and Temporal Scales

DO concentrations in the channel and in the HZ ranged between 1.28 and 5.76 mg L–1 with an overall decreasing trend from hour 0 to hour 72 (Figure 1a) and with depth (Figure 1b). Analysis for multivariate differences among all molecular peaks regardless of formula assignment showed broad compositional differences in DOM pools among the channel, DW, and UW spatial domains (p = 0.005, PERMANOVA, Figure S8). This may indicate progressive transformation of DOM as it moved through the system, due to possible partial and/or total processing of individual organic molecules through microbial metabolism. These results are consistent with previous work showing that longer HZ flowpaths (corresponding to longer residence time) enable more microbial utilization of incoming solutes, which in turn leads to a decrease in DO.11,27,57 Progressive transformation of DOM also supports residence time as a mechanism responsible for differences within the HZ and along hyporheic flowpaths (DW/UW) and the channel. Overall, across all domains, ∼71% of all observed molecules were assigned to a chemical class (i.e., ∼29% chemical class “None”). The majority of molecules with an assigned chemical class were classified as lignin-like (∼60%; Supporting Information, Tables S6–S8), which is expected considering that leaf leachate was the source of DOM in our flume and lignin-like molecules are characteristic of vascular plants.73

Figure 1.

Figure 1

Open in a new tab

Spatiotemporal dynamics of DO. (a) Planar optode images of a bedform capturing the DO field in the channel and in the hyporheic zone before the addition of leaves (hour 0) and after the removal of the leaves from the channel (hours 24 and 72). (b) Local, average DO concentrations for the channel (CH), downwelling (DW), and upwelling (UW) parts of the hyporheic zone at hour 0, 24, and 72. Analysis for temporal shifts in DO means within DW and UW groups did not result in statistically significant differences (p = 0.224 at DW, and p = 0.078 at UW, ANOVA). The sampling points and their location within the hyporheic zone flow field inferred from previous modeling studies of similar bedforms are shown on the rightmost part of (b).

A more detailed examination of DOM temporal dynamics revealed distinct changes in various chemical classes of DOM across each spatial domain (i.e., DW, UW, and the channel). Focusing on molecular peaks that did not repeat through time within a given spatial domain (i.e., nonubiquitous molecules, Supporting Information), we found similar temporal trends in DOM molecular classes between UW and the channel spatial domains and diverging trends between DW and UW spatial domains. Specifically, we note that within DW, chemical classes associated with complex molecules (i.e., lignin-, lipid-, protein-, and condensed hydrocarbon-like) tended to contain more nonubiquitous molecules through time, while UW and the channel experienced a decrease in nonubiquitous molecules of these classes through time (Figure 2a; for overlapping peaks see Figure S4).

Figure 2.

Figure 2

Open in a new tab

Temporal variations in chemical classes and thermodynamic favorability of DOM in the channel (CH), downwelling (DW), and upwelling (UW) spatial domains of the flume. (a) Spatiotemporal dynamics of DOM classes for CH, DW, and UW spatial domains represented through changes in total nonubiquitous peaks (molecules) assigned per class. Green and red arrows indicate temporal trends of increase and decrease in the total number of nonubiquitous molecules assigned to a specific chemical class within each spatial domain. Classes with a relatively unchanging count of total nonubiquitous peaks through time were left arrowless. (b) Temperature-adjusted average Gibbs free energy of carbon oxidation (ΔGCox) before the addition and after the removal of leaves. Temporal shift in ΔGCox within DW and UW groups was confirmed with ANOVA + Tukey tests (ANOVA: p = 2.8 × 10–9 at DW, p = 4.4 × 10–9 at UW, pairwise Tukey p < 0.001 for all pairs in DW and UW groups).

In interpreting temporal trends in DOM composition among different spatial domains, we emphasize the complexity of flow through the HZ. Residence time increases along the flowpaths and with depth. As such, the residence time in the UW part of the HZ exceeds the residence time in the DW part of the HZ, and the residence time of deep flowpaths exceeds the residence time of shallow flowpaths within a given HZ volume (Figures 1b and S1c). This implies that at any given time, the water sampled from a UW location close to the surface will aggregate the molecular character of DOM pools from short, shallow flowpaths (i.e., short residence time) and long, deep flowpaths (i.e., long residence time). Because of this hydrology, we hypothesize that DOM pool composition in UW should reflect both the degradation of relatively simple molecular structures (e.g., amino sugar-like), likely cycled quickly74 along short flowpaths, as well as the partial-to-complete degradation of complex molecular structures with slower decomposition rates along both short and long flowpaths. We also hypothesize that the similarities in temporal trends between chemical classes in UW and the channel could indicate the impact of HZ processes on surface water DOM pool composition, as metabolized DOM exiting the HZ was broadly similar to channel water DOM. Possible differences in biomass between DW and UW domains or effects of desorption paired with low decomposition rates over comparatively short residence time could have contributed to the observed increase in complex organic molecules through time in DW (i.e., condensed hydrocarbon-, lignin-, lipid-, and protein-like).

In addition to patterns in temporal changes in DOM pool composition, we found changes in the temperature-adjusted average Gibbs free energy of C oxidation (ΔGCox) through time, where more negative ΔGCox values indicate increased thermodynamic favorability (Figure 2b). We found that the ΔGCox in all spatial domains significantly increased through time by ∼2 kJ/mol-C (p < 0.001, ANOVA + Tukey, Figure 2b). While these results are consistent with some previous reports of increasing thermodynamic favorability over time and with enhanced respiration in the HZ,23,35 they deviate from a classic paradigm that the thermodynamic favorability of the entire DOM pool should become lower over time.75−77 One possible explanation for our results is that the comparatively high temporal resolution of sampling allowed us to observe a gradual microbial breakdown of DOM, where complex, long-chained, or aromatic polymers (e.g., lignin and some proteins, low favorability) are cleaved into smaller polymers (e.g., amino sugars, high favorability),78 with an overall lower ΔGCox and higher thermodynamic favorability. It is likely that given sufficient time, the thermodynamic favorability of DOM pools may have succumbed to follow the classic paradigm.

We also found that the C/N ratio in the flume was heterogeneous through space and time, exhibiting both vertical and lateral gradients in the HZ (Figure 3a and Table S1). When using elemental ratios to infer nutrient limitations, C limitation is associated with a C/N ratio < 5.38,39 Using this estimate as a guideline, all spatial domains of the flume appeared to be more C-limited than N-limited before the addition of leaves (hour 0, C/N between 2.48 and 3.32 in the channel and in DW, and between 2.71 and 4.10 in UW), except for DW at a 60 cm depth, which was slightly more N-limited (C/N = 5.29) (Table S1). Through time, all spatial domains shifted toward more N-limiting conditions with the largest shift at a 60 cm depth in the UW (C/N 4.10 to 6.03 from hours 0 to 72; Figure 3a; Table S1). This shift in the C/N ratio toward N limitation through time and with depth could be associated with desorption of complex DOM molecules that take longer to degrade or with preferential consumption of N-containing DOM with depth. As a result, the system should experience an enrichment in C concentration (relative to N) through space and time. Both C (as NPOC) accumulation and N (as TN) depletion through space and time were observed in our experiment (Figure 3b,c).

Figure 3.

Figure 3

Open in a new tab

Spatiotemporal dynamics of C/N ratios, NPOC, and TN. Channel (CH), downwelling (DW), and upwelling (UW) sampling domains are depicted through yellow, blue, and red color markers, respectively. The solid green line marks inferred C and N limitations.38,39 (a) C/N ratios through time (hours 0, 24, and 72) and depth in the CH, DW, and UW spatial domains. (b) Concentration of NPOC with depth and through time at DW and UW spatial domains. The results of the linear regression analysis where the change in NPOC concentrations with depth were significant (i.e., p < 0.05 are shown with black lines). (c) Same as (b) but for TN.

Based on these results, we propose that C and N limitations are dynamic, exist simultaneously across small spatial domains in the HZ, and affect AR. We suggest that shorter hyporheic flowpaths are more likely to experience C limitation, whereas longer flowpaths may be more N-limited. This is evidenced by increasing the C/N ratio through depth and with time, accumulation of NPOC, and reduction in TN below 30 cm in the HZ (Figure 3). Also, we highlight that variations in factors limiting AR can operate at smaller/shorter scales, typically aggregated in field measurements, and suggest that these dynamics are unrepresented in our current models of river corridor biogeochemical cycles.

Relationships between DO, DOM Thermodynamics, and Organic N Suggest Variation in the Regulation of AR across Small Spatial Domains

To further evaluate if AR at small measurement scales could be associated with both DOM thermodynamics and organic N in the HZ, we examined relationships between DO, DOM thermodynamics (ΔGCox), and N-containing inferred biochemical reactions (i.e., pairwise mass differences (m/z) among FTICR-MS peaks (see the Supporting Information)). We delineated each spatial domain of the HZ (i.e., UW and DW) into two chemically distinct vertical sections—shallow (10–30 cm) and deep (40–60 cm)—based on changes in concentrations of TN and NPOC with depth (Figure 3b,c). The shallow HZ was characterized by decreasing NPOC and relatively stable TN with depth in both DW and UW. The deep HZ was characterized by increasing NPOC and decreasing TN with depth in both DW and UW.

We found sequential, flowpath-delineated relationships between DO and DOM chemistry that suggest simultaneous thermodynamic and N regulation of AR over spatial and temporal scales (centimeter and hourly scales). The thermodynamic favorability of DOM was associated with DO decreases in the shallow HZ, whereas in the deep HZ, N-containing inferred biochemical reactions were the primary and positive correlate of decreasing DO (Figure 4). DO and ΔGCox were linearly related along shallow hyporheic flowpaths—as DO decreased, the thermodynamic favorability increased (more negative ΔGCox, DW: p = 0.016, R2 = 0.58; UW: p = 0.025, R2 = 0.53, Figure 4a)—but they were not significantly related in the deep HZ (DW: p = 0.249; UW: p = 0.066, Figure 4a). The relationship between thermodynamic favorability and DO in the shallow HZ, under relatively low C/N ratios, suggests that AR may be regulated by DOM thermodynamics under C limitation,35 while the lack of a relationship in the deep HZ under higher C/N suggests another limiting factor.

Figure 4.

Figure 4

Open in a new tab

Spatial relationships between DO, average Gibbs free energy of the half-reaction of organic carbon oxidation (ΔGCox), and relative abundance of N-containing inferred biochemical reactions in the HZ. Downwelling (DW) and upwelling (UW) spatial domains are depicted through blue and red colors, respectively. (a) Regression analysis for DO concentrations and ΔGCox measurements grouped by the sampling domain (e.g., DW and UW) and depth within the HZ. (Top row) DO concentrations and ΔGCox correlation for the shallow HZ (10–30 cm). (Bottom row) DO and ΔGCox correlation for the deep HZ (40–50 cm). The 60 cm depth sampling location was not included in the linear regression. (b) Linear regression analysis for DO concentration and relative abundance of inferred N-containing biochemical reactions. Top and bottom panels delineate shallow and deep HZs, respectively. The abundance of inferred N-containing biochemical reactions is listed in Table S4.

The existence of spatially delineated relationships between the DO and inferred biochemical reactions involving organic N in the deep HZ also supports our theory that AR in the HZ can be simultaneously limited by C and N availability within a small spatial domain. We found a significant and positive relationship between DO and relative abundance of inferred biochemical reactions containing N in both DW and UW spatial domains in the deep HZ (DW: p = 0.00013, R2 = 0.98; UW: p = 0.033, R2 = 0.72, Figure 4b). No significant relationship between DO and inferred N-containing biochemical reactions was found in the shallow HZ (DW: p = 0.445; UW: p = 0.067, Figure 4b).

Also, overall, we found that N was involved in 7.87% of the inferred biochemical reactions identified across all 39 samples. Approximately 4% of N-containing inferred biochemical reactions involved amino acids (Supporting Information), which are comparatively bioreactive components of DOM.74 Additionally, nitrogenous substitution involving ammonia (H3O_1N1), likely related to the process of deamination (Figure S9), was the most abundant across all samples (1.7% of inferred N-containing biochemical reactions; Table S5). This suggests the gain of ammonia (NH3) from organic molecules, which then could be oxidized to nitrite (NO2–) for energetic gain by microbes or used for biomass or amino acid production.

Microbial utilization of N gained from organic molecules under depleted N conditions is consistent with N mining theory.79,80 Nitrogen mining proposes that N limitation promotes microbial N scavenging from sources of organic N like proteins80 and amino acids81 to support the respiration of more chemically complex DOM compounds.26,82,83 In contrast, under N-replete conditions, microorganisms tend to utilize more bioavailable and thermodynamically favorable organic matter for respiration. Consequently, a transition from decomposition of bioavailable DOM to N mining can result in reduced rates of DOM decomposition.84,85 Nitrogen mining has been previously proposed as a mechanism responsible for increases in reactions involving organic N in the HZ.21,23,24

Conceptual Model for Simultaneous Regulation of AR in HZs

Recent advances in understanding the impacts of organic matter chemistry on microbially driven reactions have promoted an evolution of process-based models. In turn, biogeochemical models have progressed from using single values as proxies for complex substrates to using lumped enzymatic and microbial activities86−89 and to substrate-explicit and metabolic models that can represent chemical reactions in finer detail.53,90 Recently, the lack of a spatially and temporally explicit understanding of the evolution and processes associated with DOM pool properties has been proposed as an additional source of uncertainties in predictive models.53 For example, if AR in HZs is regulated by both DOM thermodynamics and N content of DOM under specific C/N conditions, and if this regulation is consistent with residence time, then the outcomes of model predictions may be sensitive to spatiotemporal changes in combined quantity and quality of DOM and N. Because of the complexity, high uncertainty, and broad impact of HZs on biogeochemical cycles,1,9,91 we must have solid conceptual models based on most representative measurement scales upon which to build realistic numerical scenarios.

While multiple limitations of DOM thermodynamics and N on AR within a bedform-size HZ have not been reported, we note that some investigations have suggested the importance of N availability and/or DOM thermodynamics in regulating HZ metabolism at a single point in space or time.23,24,35 Although future studies with increased temporal and spatial resolution are needed to better understand DOM metabolism across the terrestrial–aquatic continuum, we find encouraging overarching agreement between the results from this and past studies with low spatiotemporal resolution into the HZ. Therefore, through our observations, we propose that distributed C/N conditions could generate selective use of DOM based on its thermodynamic properties or N content over spatial and temporal scales (centimeter and hourly scales) that currently aggregate into single measurements of AR (Figure 5). We suggest that HZs are likely to have both C and nutrient limitations present along and across very short hyporheic flowpaths, even across well-oxygenated environments. As the residence time increases along the flowpath (DW to UW), and with depth (shallow to deep HZ), the C/N ratio gradually increases shifting from C-limiting to N-limiting conditions. This gradual transition in the C/N ratio is likely to affect microbial metabolism within the HZ such that under C-replete conditions AR is oriented toward metabolism of the most bioavailable, thermodynamically favorable C molecules, whereas under N-limitation, AR is dependent on N content of organic molecules. This selective metabolism of DOM under changing C and N stoichiometry is likely to affect not only the understanding of the function of the HZ as a biochemical reactor but also the ability to predict HZ’s capacity to store C. Our results present an opportunity to refine the existing process-based model structures that are informed by measurements taken at relatively coarse spatial and temporal scales and allow the consideration of more realistic environmental scenarios.

Figure 5.

Figure 5

Open in a new tab

Conceptual model for simultaneous and spatially distributed regulation of AR in HZs. HZs have nonuniformly distributed nutrient-limiting conditions with both C and N limitations simultaneously present in different parts of the HZ, which could induce preferential uptake of DOM based on its thermodynamic favorability or N content. In the shallow HZ with short flowpaths and likely prevalent C-limiting conditions (low C/N ratios), AR relates to thermodynamic favorability of DOM molecules, whereas in the deep HZ with longer flowpaths and likely N-limiting conditions (high C/N ratios), AR relates to N content of DOM molecules.

Acknowledgments

This research was supported by the U.S. Department of Energy (DOE), Office of Science, Office of Biological and Environmental Research (BER), Environmental System Science (ESS) Program, with contributions originating from the River Corridor Scientific Focus Area (SFA) project at Pacific Northwest National Laboratory (PNNL). PNNL is operated by Battelle Memorial Institute for DOE under Contract DE-AC05-76RL01830. Also, this research was supported by the DOE, BER as part of BER’s Subsurface Biogeochemical Research (SBR) Program, with contributions originating from Grant DE-SC0018042 to The University of Texas and Pennsylvania State University and in partnership with the SFA at the Pacific Northwest National Laboratory (PNNL). Additionally, this research was supported by the DOE, Office of Science, Office of Workforce Development for Teachers and Scientists, Office of Science Graduate Student Research (SCGSR) program. The SCGSR program is administered by the Oak Ridge Institute for Science and Education for the DOE under contract number DE-SC0014664. Also, a portion of the research was performed under user proposal 51750 at Environmental Molecular Sciences Laboratory (EMSL), a DOE Office of Science User Facility sponsored by the BER. Additionally, we thank Pawel M. Wojtowicz for help in the laboratory and in the field and Lupita Renteria and Jacqueline Wells for help with FTICR-MS laboratory prep work.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.est.3c04765.

  • Reference list for biochemical reactions (XLSX)

  • Abundance of each N reaction across the data set (XLSX)

  • Additional text regarding Formularity, identification of nonubiquitous molecules, the calculation of Gibbs free energy (ΔGCox), identification of inferred biochemical reactions and nitrogen (N)-containing inferred biochemical reactions, and calculations of relative abundances of chemical classes; figures showing laboratory setup and equipment, settings used in Formularity software, plots showing shared peaks and nonubiquitous molecules in each domain through time, distributions of dissolved oxygen and Gibbs free energy, schematic for identification of biochemical reactions, NMDS analysis for broad compositional differences of the DOM pool, and the process of deamination; tables showing concentrations of NPOC, TN, pH, C/N ratios, temperature, relative abundances of N-containing inferred biochemical reactions, and relative abundance of chemical classes through time (PDF)

Author Present Address

# Pacific Northwest National Laboratory, Richland, Washington 99352, United States (A.B.T.)

Author Contributions

A.B.T., V.A.G-C., E.B.G., M.H.K., and M.B.C. conceptualized the study; A.B.T. conducted sample collection; A.B.T., V.A.G-C., and R.E.D. completed data analysis; A.B.T., V.A.G-C., and E.B.G. drafted the manuscript; and all authors contributed to editing the manuscript. All authors have read and agreed to the published version of the manuscript.

The authors declare no competing financial interest.

Supplementary Material

es3c04765_si_001.xlsx (40.1KB, xlsx)
es3c04765_si_002.xlsx (22.5KB, xlsx)
es3c04765_si_003.pdf (1.7MB, pdf)

References

  1. Battin T. J.; Luyssaert S.; Kaplan L. A.; Aufdenkampe A. K.; Richter A.; Tranvik L. J. The Boundless Carbon Cycle. Nat. Geosci. 2009, 2, 598–600. 10.1038/ngeo618. [DOI] [Google Scholar]
  2. Seitzinger S.; Harrison J. A.; Böhlke J. K.; Bouwman A. F.; Lowrance R.; Peterson B.; Tobias C.; Drecht G. V. Denitrification Across Landscapes and Waterscapes: A Synthesis. Ecol. Appl. 2006, 16 (6), 2064–2090. 10.1890/1051-0761(2006)016[2064:DALAWA]2.0.CO;2. [DOI] [PubMed] [Google Scholar]
  3. Wollheim W. M.; Vörösmarty C. J.; Bouwman A. F.; Green P.; Harrison J.; Linder E.; Peterson B. J.; Seitzinger S. P.; Syvitski J. P. M. Global N Removal by Freshwater Aquatic Systems Using a Spatially Distributed, within-Basin Approach. Glob. Biogeochem. Cycles 2008, 22 (2), GB2026. 10.1029/2007GB002963. [DOI] [Google Scholar]
  4. Drake T. W.; Raymond P. A.; Spencer R. G. M. Terrestrial Carbon Inputs to Inland Waters: A Current Synthesis of Estimates and Uncertainty. Limnol. Oceanogr. Lett. 2018, 3 (3), 132–142. 10.1002/lol2.10055. [DOI] [Google Scholar]
  5. McDowell W. H.; Fisher S. G. Autumnal Processing of Dissolved Organic Matter in a Small Woodland Stream Ecosystem. Ecology 1976, 57 (3), 561–569. 10.2307/1936440. [DOI] [Google Scholar]
  6. Meyer J. L.; Wallace J. B.; Eggert S. L. Leaf Litter as a Source of Dissolved Organic Carbon in Streams. Ecosystems 1998, 1 (3), 240–249. 10.1007/s100219900019. [DOI] [Google Scholar]
  7. Meyer J. L.; Tate C. M. The Effects of Watershed Disturbance on Dissolved Organic Carbon Dynamics of a Stream. Ecology 1983, 64, 33–44. 10.2307/1937326. [DOI] [Google Scholar]
  8. Raymond P. A.; Hartmann J.; Lauerwald R.; Sobek S.; McDonald C.; Hoover M.; Butman D.; Striegl R.; Mayorga E.; Humborg C.; Kortelainen P.; Dürr H.; Meybeck M.; Ciais P.; Guth P. Global Carbon Dioxide Emissions from Inland Waters. Nature 2013, 503 (7476), 355–359. 10.1038/nature12760. [DOI] [PubMed] [Google Scholar]
  9. Boulton A. J.; Findlay S.; Marmonier P.; Stanley E. H.; Valett H. M. The Functional Significance of the Hyporheic Zone in Streams and Rivers. Annu. Rev. Ecol. Syst. 1998, 29, 59–81. 10.1146/annurev.ecolsys.29.1.59. [DOI] [Google Scholar]
  10. Gomez-Velez J. D.; Harvey J. W.; Cardenas M. B.; Kiel B. Denitrification in the Mississippi River Network Controlled by Flow through River Bedforms. Nat. Geosci. 2015, 8 (12), 941–945. 10.1038/ngeo2567. [DOI] [Google Scholar]
  11. Krause S.; Hannah D. M.; Fleckenstein J. H.; Heppell C. M.; Kaeser D.; Pickup R.; Pinay G.; Robertson A. L.; Wood P. J. Inter-Disciplinary Perspectives on Processes in the Hyporheic Zone. Ecohydrology 2011, 4 (4), 481–499. 10.1002/eco.176. [DOI] [Google Scholar]
  12. Stegen J. C.; Fredrickson J. K.; Wilkins M. J.; Konopka A. E.; Nelson W. C.; Arntzen E. V.; Chrisler W. B.; Chu R. K.; Danczak R. E.; Fansler S. J.; Kennedy D. W.; Resch C. T.; Tfaily M. Groundwater–Surface Water Mixing Shifts Ecological Assembly Processes and Stimulates Organic Carbon Turnover. Nat. Commun. 2016, 7 (1), 11237. 10.1038/ncomms11237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Naegeli M. W.; Uehlinger U. Contribution of the Hyporheic Zone to Ecosystem Metabolism in a Prealpine Gravel-Bed-River. J. North Am. Benthol. Soc. 1997, 16 (4), 794–804. 10.2307/1468172. [DOI] [Google Scholar]
  14. Berggren M.; del Giorgio P. A. Distinct Patterns of Microbial Metabolism Associated to Riverine Dissolved Organic Carbon of Different Source and Quality. J. Geophys. Res.: Biogeosci. 2015, 120 (6), 989–999. 10.1002/2015JG002963. [DOI] [Google Scholar]
  15. Bianchi T. S. The Role of Terrestrially Derived Organic Carbon in the Coastal Ocean: A Changing Paradigm and the Priming Effect. Proc. Natl. Acad. Sci. U. S. A. 2011, 108 (49), 19473–19481. 10.1073/pnas.1017982108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cory R. M.; Kaplan L. A. Biological Lability of Streamwater Fluorescent Dissolved Organic Matter. Limnol. Oceanogr. 2012, 57 (5), 1347–1360. 10.4319/lo.2012.57.5.1347. [DOI] [Google Scholar]
  17. Koehler B.; von Wachenfeldt E.; Kothawala D.; Tranvik L. J. Reactivity Continuum of Dissolved Organic Carbon Decomposition in Lake Water. J. Geophys. Res.: Biogeosci. 2012, 117 (G1), 001793 10.1029/2011JG001793. [DOI] [Google Scholar]
  18. Shen Y.; Chapelle F.; Strom E.; Benner R. Origins and Bioavailability of Dissolved Organic Matter in Groundwater. Biogeochemistry 2015, 122 (1), 61–78. 10.1007/s10533-014-0029-4. [DOI] [Google Scholar]
  19. Tfaily M. M.; Chu R. K.; Toyoda J.; Tolić N.; Robinson E. W.; Paša-Tolić L.; Hess N. J. Sequential Extraction Protocol for Organic Matter from Soils and Sediments Using High Resolution Mass Spectrometry. Anal. Chim. Acta 2017, 972, 54–61. 10.1016/j.aca.2017.03.031. [DOI] [PubMed] [Google Scholar]
  20. Tremblay L. B.; Dittmar T.; Marshall A. G.; Cooper W. J.; Cooper W. T. Molecular Characterization of Dissolved Organic Matter in a North Brazilian Mangrove Porewater and Mangrove-Fringed Estuaries by Ultrahigh Resolution Fourier Transform-Ion Cyclotron Resonance Mass Spectrometry and Excitation/Emission Spectroscopy. Mar. Chem. 2007, 105 (1), 15–29. 10.1016/j.marchem.2006.12.015. [DOI] [Google Scholar]
  21. Danczak R. E.; Goldman A. E.; Chu R. K.; Toyoda J. G.; Garayburu-Caruso V. A.; Tolić N.; Graham E. B.; Morad J. W.; Renteria L.; Wells J. R.; Herzog S. P.; Ward A. S.; Stegen J. C. Ecological Theory Applied to Environmental Metabolomes Reveals Compositional Divergence despite Conserved Molecular Properties. Sci. Total Environ. 2021, 788, 147409 10.1016/j.scitotenv.2021.147409. [DOI] [PubMed] [Google Scholar]
  22. Graham E. B.; Tfaily M. M.; Crump A. R.; Goldman A. E.; Bramer L. M.; Arntzen E.; Romero E.; Resch C. T.; Kennedy D. W.; Stegen J. C. Carbon Inputs From Riparian Vegetation Limit Oxidation of Physically Bound Organic Carbon Via Biochemical and Thermodynamic Processes. J. Geophys. Res. Biogeosci. 2017, 122 (12), 3188–3205. 10.1002/2017JG003967. [DOI] [Google Scholar]
  23. Stegen J. C.; Johnson T.; Fredrickson J. K.; Wilkins M. J.; Konopka A. E.; Nelson W. C.; Arntzen E. V.; Chrisler W. B.; Chu R. K.; Fansler S. J.; Graham E. B.; Kennedy D. W.; Resch C. T.; Tfaily M.; Zachara J. Influences of Organic Carbon Speciation on Hyporheic Corridor Biogeochemistry and Microbial Ecology. Nat. Commun. 2018, 9 (1), 585. 10.1038/s41467-018-02922-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Graham E. B.; Crump A. R.; Kennedy D. W.; Arntzen E.; Fansler S.; Purvine S. O.; Nicora C. D.; Nelson W.; Tfaily M. M.; Stegen J. C. Multi’omics Comparison Reveals Metabolome Biochemistry, Not Microbiome Composition or Gene Expression, Corresponds to Elevated Biogeochemical Function in the Hyporheic Zone. Sci. Total Environ. 2018, 642, 742–753. 10.1016/j.scitotenv.2018.05.256. [DOI] [PubMed] [Google Scholar]
  25. Berg B.; McClaugherty C.. Plant Litter: Decomposition, Humus Formation, Carbon Sequestration; Springer International Publishing: Cham, 2020. [Google Scholar]
  26. Craine J. M.; Morrow C.; Fierer N. Microbial Nitrogen Limitation Increases Decomposition. Ecology 2007, 88 (8), 2105–2113. 10.1890/06-1847.1. [DOI] [PubMed] [Google Scholar]
  27. Zarnetske J. P.; Haggerty R.; Wondzell S. M.; Baker M. A. Dynamics of Nitrate Production and Removal as a Function of Residence Time in the Hyporheic Zone. J. Geophys. Res. Biogeosci. 2011, 116 (G1), 001356 10.1029/2010JG001356. [DOI] [Google Scholar]
  28. Liu Q.; Liang Y.; Cai W.-J.; Wang K.; Wang J.; Yin K. Changing Riverine Organic C:N Ratios along the Pearl River: Implications for Estuarine and Coastal Carbon Cycles. Sci. Total Environ. 2020, 709, 136052 10.1016/j.scitotenv.2019.136052. [DOI] [PubMed] [Google Scholar]
  29. Sullivan B. W.; Alvarez-Clare S.; Castle S. C.; Porder S.; Reed S. C.; Schreeg L.; Townsend A. R.; Cleveland C. C. Assessing Nutrient Limitation in Complex Forested Ecosystems: Alternatives to Large-Scale Fertilization Experiments. Ecology 2014, 95 (3), 668–681. 10.1890/13-0825.1. [DOI] [PubMed] [Google Scholar]
  30. Webster J. R.; Newbold J. D.; Lin L.Nutrient Spiraling and Transport in Streams: The Importance of In-Stream Biological Processes to Nutrient Dynamics in Streams. In Stream Ecosystems in a Changing Environment; Jones J. B., Stanley E. H., Eds.; Academic Press: Boston, 2016; 181–239. [Google Scholar]
  31. Manzoni S.; Jackson R. B.; Trofymow J. A.; Porporato A. The Global Stoichiometry of Litter Nitrogen Mineralization. Science 2008, 321 (5889), 684–686. 10.1126/science.1159792. [DOI] [PubMed] [Google Scholar]
  32. Sterner R. W.; Elser J. J.. Ecological Stoichiometry; Princeton University Press, 2002. [Google Scholar]
  33. Elser J. J.; Sterner R. W.; Gorokhova E.; Fagan W. F.; Markow T. A.; Cotner J. B.; Harrison J. F.; Hobbie S. E.; Odell G. M.; Weider L. W. Biological Stoichiometry from Genes to Ecosystems. Ecol. Lett. 2000, 3 (6), 540–550. 10.1111/j.1461-0248.2000.00185.x. [DOI] [Google Scholar]
  34. Graham E. B.; Hofmockel K. S. Ecological Stoichiometry as a Foundation for Omics-Enabled Biogeochemical Models of Soil Organic Matter Decomposition. Biogeochemistry 2022, 157 (1), 31–50. 10.1007/s10533-021-00851-2. [DOI] [Google Scholar]
  35. Garayburu-Caruso V. A.; Stegen J. C.; Song H.-S.; Renteria L.; Wells J.; Garcia W.; Resch C. T.; Goldman A. E.; Chu R. K.; Toyoda J.; Graham E. B. Carbon Limitation Leads to Thermodynamic Regulation of Aerobic Metabolism. Environ. Sci. Technol. Lett. 2020, 7 (7), 517–524. 10.1021/acs.estlett.0c00258. [DOI] [Google Scholar]
  36. Cory R. M.; Kling G. W. Interactions between Sunlight and Microorganisms Influence Dissolved Organic Matter Degradation along the Aquatic Continuum. Limnol. Oceanogr. Lett. 2018, 3 (3), 102–116. 10.1002/lol2.10060. [DOI] [Google Scholar]
  37. Hill B. H.; Elonen C. M.; Seifert L. R.; May A. A.; Tarquinio E. Microbial Enzyme Stoichiometry and Nutrient Limitation in US Streams and Rivers. Ecol. Indic. 2012, 18, 540–551. 10.1016/j.ecolind.2012.01.007. [DOI] [Google Scholar]
  38. Egli T.; Lendenmann U.; Snozzi M. Kinetics of Microbial Growth with Mixtures of Carbon Sources. Antonie van Leeuwenhoek 1993, 63 (3), 289–298. 10.1007/BF00871224. [DOI] [PubMed] [Google Scholar]
  39. Zinn M.; Witholt B.; Egli T. Dual Nutrient Limited Growth: Models, Experimental Observations, and Applications. J. Biotechnol. 2004, 113 (1), 263–279. 10.1016/j.jbiotec.2004.03.030. [DOI] [PubMed] [Google Scholar]
  40. Dittmar T.; Koch B.; Hertkorn N.; Kattner G. A Simple and Efficient Method for the Solid-Phase Extraction of Dissolved Organic Matter (SPE-DOM) from Seawater. Limnol. Oceanogr.: Methods 2008, 6 (6), 230–235. 10.4319/lom.2008.6.230. [DOI] [Google Scholar]
  41. Garayburu-Caruso V. A.; Danczak R. E.; Stegen J. C.; Renteria L.; Mccall M.; Goldman A. E.; Chu R. K.; Toyoda J.; Resch C. T.; Torgeson J. M.; Wells J.; Fansler S.; Kumar S.; Graham E. B. Using Community Science to Reveal the Global Chemogeography of River Metabolomes. Metabolites 2020, 10 (12), 518. 10.3390/metabo10120518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Tolić N.; Liu Y.; Liyu A.; Shen Y.; Tfaily M. M.; Kujawinski E. B.; Longnecker K.; Kuo L.-J.; Robinson E. W.; Paša-Tolić L.; Hess N. J. Formularity: Software for Automated Formula Assignment of Natural and Other Organic Matter from Ultrahigh-Resolution Mass Spectra. Anal. Chem. 2017, 89, 12659–12665. 10.1021/acs.analchem.7b03318. [DOI] [PubMed] [Google Scholar]
  43. Bramer L. M.; White A.. FtmsRanalysis: Analysis and Visualization Tools for FT-MS Data, 2019. https://emsl-computing.github.io/ftmsRanalysis/.
  44. Bramer L. M.; White A. M.; Stratton K. G.; Thompson A. M.; Claiborne D.; Hofmockel K.; McCue L. A. FtmsRanalysis: An R Package for Exploratory Data Analysis and Interactive Visualization of FT-MS Data. PLoS Comput. Biol. 2020, 16 (3), e1007654 10.1371/journal.pcbi.1007654. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kim S.; Kramer R. W.; Hatcher P. G. Graphical Method for Analysis of Ultrahigh-Resolution Broadband Mass Spectra of Natural Organic Matter, the Van Krevelen Diagram. Anal. Chem. 2003, 75 (20), 5336–5344. 10.1021/ac034415p. [DOI] [PubMed] [Google Scholar]
  46. Tfaily M. M.; Chu R. K.; Tolić N.; Roscioli K. M.; Anderton C. R.; Paša-Tolić L.; Robinson E. W.; Hess N. J. Advanced Solvent Based Methods for Molecular Characterization of Soil Organic Matter by High-Resolution Mass Spectrometry. Anal. Chem. 2015, 87 (10), 5206–5215. 10.1021/acs.analchem.5b00116. [DOI] [PubMed] [Google Scholar]
  47. Kew W.; Myers-Pigg A.; Chang C.; Colby S.; Eder J.; Tfaily M.; Hawkes J.; Chu R.; Stegen J.. Reviews and Syntheses: Use and Misuse of Peak Intensities from High Resolution Mass Spectrometry in Organic Matter Studies: Opportunities for Robust Usage; preprint; Biogeochem.: Org. Biogeochem., 2022. 10.5194/egusphere-2022-1105. [DOI] [Google Scholar]
  48. Kujawinski E. B.; Behn M. D. Automated Analysis of Electrospray Ionization Fourier Transform Ion Cyclotron Resonance Mass Spectra of Natural Organic Matter. Anal. Chem. 2006, 78 (13), 4363–4373. 10.1021/ac0600306. [DOI] [PubMed] [Google Scholar]
  49. Minor E. C.; Steinbring C. J.; Longnecker K.; Kujawinski E. B. Characterization of Dissolved Organic Matter in Lake Superior and Its Watershed Using Ultrahigh Resolution Mass Spectrometry. Org. Geochem. 2012, 43, 1–11. 10.1016/j.orggeochem.2011.11.007. [DOI] [Google Scholar]
  50. Bahureksa W.; Tfaily M. M.; Boiteau R. M.; Young R. B.; Logan M. N.; McKenna A. M.; Borch T. Soil Organic Matter Characterization by Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FTICR MS): A Critical Review of Sample Preparation, Analysis, and Data Interpretation. Environ. Sci. Technol. 2021, 55 (14), 9637–9656. 10.1021/acs.est.1c01135. [DOI] [PubMed] [Google Scholar]
  51. Cates A. M.; Jilling A.; Tfaily M. M.; Jackson R. D. Temperature and Moisture Alter Organic Matter Composition across Soil Fractions. Geoderma 2022, 409, 115628 10.1016/j.geoderma.2021.115628. [DOI] [Google Scholar]
  52. LaRowe D. E.; Van Cappellen P. Degradation of Natural Organic Matter: A Thermodynamic Analysis. Geochim. Cosmochim. Acta 2011, 75 (8), 2030–2042. 10.1016/j.gca.2011.01.020. [DOI] [Google Scholar]
  53. Song H.-S.; Stegen J. C.; Graham E. B.; Lee J.-Y.; Garayburu-Caruso V. A.; Nelson W. C.; Chen X.; Moulton J. D.; Scheibe T. D. Representing Organic Matter Thermodynamics in Biogeochemical Reactions via Substrate-Explicit Modeling. Front. Microbiol. 2020, 11, 531756 10.3389/fmicb.2020.531756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Bailey V. L.; Smith A. P.; Tfaily M.; Fansler S. J.; Bond-Lamberty B. Differences in Soluble Organic Carbon Chemistry in Pore Waters Sampled from Different Pore Size Domains. Soil Biol. Biochem. 2017, 107, 133–143. 10.1016/j.soilbio.2016.11.025. [DOI] [Google Scholar]
  55. Breitling R.; Ritchie S.; Goodenowe D.; Stewart M. L.; Barrett M. P. Ab Initio Prediction of Metabolic Networks Using Fourier Transform Mass Spectrometry Data. Metabolomics 2006, 2 (3), 155–164. 10.1007/s11306-006-0029-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Sengupta A.; Indivero J.; Gunn C.; Tfaily M. M.; Chu R. K.; Toyoda J.; Bailey V. L.; Ward N. D.; Stegen J. C. Spatial Gradients in the Characteristics of Soil-Carbon Fractions Are Associated with Abiotic Features but Not Microbial Communities. Biogeosciences 2019, 16 (19), 3911–3928. 10.5194/bg-16-3911-2019. [DOI] [Google Scholar]
  57. Kaufman M. H.; Cardenas M. B.; Buttles J.; Kessler A. J.; Cook P. L. M. Hyporheic Hot Moments: Dissolved Oxygen Dynamics in the Hyporheic Zone in Response to Surface Flow Perturbations. Water Resour. Res. 2017, 53 (8), 6642–6662. 10.1002/2016WR020296. [DOI] [Google Scholar]
  58. Larsen M.; Borisov S. M.; Grunwald B.; Klimant I.; Glud R. N. A Simple and Inexpensive High Resolution Color Ratiometric Planar Optode Imaging Approach: Application to Oxygen and PH Sensing. Limnol. Oceanogr. Methods 2011, 9 (9), 348–360. 10.4319/lom.2011.9.348. [DOI] [Google Scholar]
  59. Van Rossum G.; Drake F. L.. Python 3 Reference Manual; CreateSpace: Scotts Valley, CA, 2009. [Google Scholar]
  60. Clark J.; Pillow A.. Pillow (PIL Fork). https://pillow.readthedocs.io/en/stable/index.html (accessed 2023–08–09).
  61. Harris C. R.; Millman K. J.; van der Walt S. J.; Gommers R.; Virtanen P.; Cournapeau D.; Wieser E.; Taylor J.; Berg S.; Smith N. J.; Kern R.; Picus M.; Hoyer S.; van Kerkwijk M. H.; Brett M.; Haldane A.; del Río J. F.; Wiebe M.; Peterson P.; Gérard-Marchant P.; Sheppard K.; Reddy T.; Weckesser W.; Abbasi H.; Gohlke C.; Oliphant T. E. Array Programming with NumPy. Nature 2020, 585 (7825), 357–362. 10.1038/s41586-020-2649-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Borisov S. M.; Vasylevska A. S.; Krause Ch.; Wolfbeis O. S. Composite Luminescent Material for Dual Sensing of Oxygen and Temperature. Adv. Funct. Mater. 2006, 16 (12), 1536–1542. 10.1002/adfm.200500778. [DOI] [Google Scholar]
  63. Team R. C.R: The R Project for Statistical Computing. https://www.r-project.org/ (accessed 2023–08–09).
  64. Oksanen J.; Simpson G. L.; Blanchet F. G.; Kindt R.; Legendre P.; Minchin P. R.; O’Hara R. B.; Solymos P.; Stevens M. H. H.; Szoecs E.; Wagner H.; Barbour M.; Bedward M.; Bolker B.; Borcard D.; Carvalho G.; Chirico M.; Caceres M. D.; Durand S.; Evangelista H. B. A.; FitzJohn R.; Friendly M.; Furneaux B.; Hannigan G.; Hill M. O.; Lahti L.; McGlinn D.; Ouellette M.-H.; Cunha E. R.; Smith T.; Stier A.; Braak C. J. F. T.; Weedon J.. Vegan: Community Ecology Package, 2022. https://cran.r-project.org/web/packages/vegan/index.html (accessed 2023–08–09).
  65. Wickham H.; Chang W.; Henry L.; Pedersen T. L.; Takahashi K.; Wilke C.; Woo K.; Yutani H.; Dunnington D.; Posit PBC. Ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics, 2023. https://cran.r-project.org/web/packages/ggplot2/index.html (accessed 2023–08–09).
  66. McKinney W.Data Structures for Statistical Computing in Python. In Proceedings of the 9th Python in Science Conference; 2010; 445, 56–61.
  67. Pedregosa F.; Varoquaux G.; Gramfort A.; Michel V.; Thirion B.; Grisel O.; Blondel M.; Prettenhofer P.; Weiss R.; Dubourg V.; Vanderplas J.; Passos A.; Cournapeau D.; Brucher M.; Perrot M.; Duchesnay E. Scikit-Learn: Machine Learning in Python. J. Mach. Lean. Res. 2011, 12, 2825–2830. [Google Scholar]
  68. Virtanen P.; Gommers R.; Oliphant T. E.; Haberland M.; Reddy T.; Cournapeau D.; Burovski E.; Peterson P.; Weckesser W.; Bright J.; van der Walt S. J.; Brett M.; Wilson J.; Millman K. J.; Mayorov N.; Nelson A. R. J.; Jones E.; Kern R.; Larson E.; Carey C. J.; Polat I.; Feng Y.; Moore E. W.; Vanderplas J.; Laxalde D.; Perktold J.; Cimrman R.; Henriksen I.; Quintero E. A.; Harris C. R.; Archibald A. M.; Ribeiro A. H.; Pedregosa F.; van Mulbregt P. SciPy 1.0 Contributors. SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python. Nat. Methods 2020, 17 (3), 261–272. 10.1038/s41592-019-0686-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Camenzind T.; Hättenschwiler S.; Treseder K. K.; Lehmann A.; Rillig M. C. Nutrient Limitation of Soil Microbial Processes in Tropical Forests. Ecol. Monogr. 2018, 88 (1), 4–21. 10.1002/ecm.1279. [DOI] [Google Scholar]
  70. Sinsabaugh R. L.; Follstad Shah J. J. Ecoenzymatic Stoichiometry and Ecological Theory. Annu. Rev. Ecol. Evol. Syst. 2012, 43 (1), 313–343. 10.1146/annurev-ecolsys-071112-124414. [DOI] [Google Scholar]
  71. Mack M. C.; Schuur E. A. G.; Bret-Harte M. S.; Shaver G. R.; Chapin F. S. Ecosystem Carbon Storage in Arctic Tundra Reduced by Long-Term Nutrient Fertilization. Nature 2004, 431 (7007), 440–443. 10.1038/nature02887. [DOI] [PubMed] [Google Scholar]
  72. Hessen D. O.; Ågren G. I.; Anderson T. R.; Elser J. J.; de Ruiter P. C. Carbon Sequestration in Ecosystems: The Role of Stoichiometry. Ecology 2004, 85 (5), 1179–1192. 10.1890/02-0251. [DOI] [Google Scholar]
  73. Sarkanen K. V.; Ludwig C. H.. Lignins: Occurrence, Formation, Structure, and Reactions; Wiley-Interscience: New York, 1971. [Google Scholar]
  74. Bai Z.; Bodé S.; Huygens D.; Zhang X.; Boeckx P. Kinetics of Amino Sugar Formation from Organic Residues of Different Quality. Soil Biol. Biochem. 2013, 57, 814–821. 10.1016/j.soilbio.2012.08.006. [DOI] [Google Scholar]
  75. Hedin L. O.; von Fischer J. C.; Ostrom N. E.; Kennedy B. P.; Brown M. G.; Robertson G. P. Thermodynamic Constraints on Nitrogen Transformations and Other Biogeochemical Processes at Soil–Stream Interfaces. Ecology 1998, 79 (2), 684–703. 10.1890/0012-9658(1998)079[0684:TCONAO]2.0.CO;2. [DOI] [Google Scholar]
  76. Sobczak W. V.; Findlay S. Variation in Bioavailability of Dissolved Organic Carbon Among Stream Hyporheic Flowpaths. Ecology 2002, 83 (11), 3194–3209. 10.1890/0012-9658(2002)083[3194:VIBODO]2.0.CO;2. [DOI] [Google Scholar]
  77. Zarnetske J. P.; Haggerty R.; Wondzell S. M.; Baker M. A. Labile Dissolved Organic Carbon Supply Limits Hyporheic Denitrification. J. Geophys. Res. Biogeosci. 2011, 116 (G4), 001730 10.1029/2011JG001730. [DOI] [Google Scholar]
  78. Allison S. D.; Weintraub M. N.; Gartner T. B.; Waldrop M. P.. Evolutionary-Economic Principles as Regulators of Soil Enzyme Production and Ecosystem Function. In Soil Enzymology; Shukla G., Varma A., Eds.; Soil Biology; Springer: Berlin, Heidelberg, 2011; 229–243. [Google Scholar]
  79. Sims G. K. Nitrogen Starvation Promotes Biodegradation of N-Heterocyclic Compounds in Soil. Soil Biol. Biochem. 2006, 38 (8), 2478–2480. 10.1016/j.soilbio.2006.01.006. [DOI] [Google Scholar]
  80. Sims G. K.; Wander M. M. Proteolytic Activity under Nitrogen or Sulfur Limitation. Appl. Soil Ecol. 2002, 19 (3), 217–221. 10.1016/S0929-1393(01)00192-5. [DOI] [Google Scholar]
  81. Atkinson M. R.; Fisher S. H. Identification of Genes and Gene Products Whose Expression Is Activated during Nitrogen-Limited Growth in Bacillus Subtilis. J. Bacteriol. 1991, 173 (1), 23–27. 10.1128/jb.173.1.23-27.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  82. Berg B.; McClaugherty C.. Plant Litter: Decomposition, Humus Formation, Carbon Sequestration; Springer, Berlin, Heidelberg: Germany, 2008. [Google Scholar]
  83. Moorhead D. L.; Sinsabaugh R. L. A Theoretical Model of Litter Decay and Microbial Interaction. Ecol. Monogr. 2006, 76 (2), 151–174. 10.1890/0012-9615(2006)076[0151:ATMOLD]2.0.CO;2. [DOI] [Google Scholar]
  84. Weintraub M. N.; Schimel J. P. Interactions between Carbon and Nitrogen Mineralization and Soil Organic Matter Chemistry in Arctic Tundra Soils. Ecosystems 2003, 6 (2), 129–143. 10.1007/s10021-002-0124-6. [DOI] [Google Scholar]
  85. Zhang D.; Hui D.; Luo Y.; Zhou G. Rates of Litter Decomposition in Terrestrial Ecosystems: Global Patterns and Controlling Factors. J. Plant Ecol. 2008, 1 (2), 85–93. 10.1093/jpe/rtn002. [DOI] [Google Scholar]
  86. Allison S. D. A Trait-Based Approach for Modelling Microbial Litter Decomposition. Ecol. Lett. 2012, 15 (9), 1058–1070. 10.1111/j.1461-0248.2012.01807.x. [DOI] [PubMed] [Google Scholar]
  87. Blankinship J. C.; Berhe A. A.; Crow S. E.; Druhan J. L.; Heckman K. A.; Keiluweit M.; Lawrence C. R.; Marín-Spiotta E.; Plante A. F.; Rasmussen C.; Schädel C.; Schimel J. P.; Sierra C. A.; Thompson A.; Wagai R.; Wieder W. R. Improving Understanding of Soil Organic Matter Dynamics by Triangulating Theories, Measurements, and Models. Biogeochemistry 2018, 140 (1), 1–13. 10.1007/s10533-018-0478-2. [DOI] [Google Scholar]
  88. Schimel J. P.; Schaeffer S. M. Microbial Control over Carbon Cycling in Soil. Front. Microbiol. 2012, 3, 348. 10.3389/fmicb.2012.00348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  89. Wieder W. R.; Allison S. D.; Davidson E. A.; Georgiou K.; Hararuk O.; He Y.; Hopkins F.; Luo Y.; Smith M. J.; Sulman B.; Todd-Brown K.; Wang Y.-P.; Xia J.; Xu X. Explicitly Representing Soil Microbial Processes in Earth System Models. Glob. Biogeochem. Cycles 2015, 29 (10), 1782–1800. 10.1002/2015GB005188. [DOI] [Google Scholar]
  90. Zorrilla F.; Buric F.; Patil K. R.; Zelezniak A. MetaGEM: Reconstruction of Genome Scale Metabolic Models Directly from Metagenomes. Nucleic Acids Res. 2021, 49 (21), e126 10.1093/nar/gkab815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  91. Cole J.; Prairie Y.; Caraco N.; McDowell W.; Tranvik L.; Striegl R.; Duarte C.; Kortelainen P.; Downing J.; Middelburg J.; Melack J. Plumbing the Global Carbon Cycle: Integrating Inland Waters into the Terrestrial Carbon Budget. Ecosystems 2007, 10 (1), 172–185. 10.1007/s10021-006-9013-8. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

es3c04765_si_001.xlsx (40.1KB, xlsx)
es3c04765_si_002.xlsx (22.5KB, xlsx)
es3c04765_si_003.pdf (1.7MB, pdf)

Articles from Environmental Science & Technology are provided here courtesy of American Chemical Society

RESOURCES