<?xml-model href='http://www.tei-c.org/release/xml/tei/custom/schema/relaxng/tei_all.rng' schematypens='http://relaxng.org/ns/structure/1.0'?><TEI xmlns="http://www.tei-c.org/ns/1.0">
	<teiHeader>
		<fileDesc>
			<titleStmt><title level='a'>Chronic Warming and Nitrogen-Addition Alter Soil Organic Matter Molecular Composition Distinctly in Tandem Compared to Individual Stressors</title></titleStmt>
			<publicationStmt>
				<publisher></publisher>
				<date>03/16/2023</date>
			</publicationStmt>
			<sourceDesc>
				<bibl> 
					<idno type="par_id">10404002</idno>
					<idno type="doi">10.1021/acsearthspacechem.2c00380</idno>
					<title level='j'>ACS Earth and Space Chemistry</title>
<idno>2472-3452</idno>
<biblScope unit="volume">7</biblScope>
<biblScope unit="issue">3</biblScope>					

					<author>Iuliana Stoica</author><author>Maryam Tabatabaei Anaraki</author><author>Thomas Muratore</author><author>Melissa Knorr</author><author>Serita D. Frey</author><author>Myrna J. Simpson</author>
				</bibl>
			</sourceDesc>
		</fileDesc>
		<profileDesc>
			<abstract><ab><![CDATA[Forest soils are major reservoirs of carbon (C), but these stores are threatened by increasing global temperatures and atmospheric nitrogen (N) deposition. These environmental stressors can alter soil microbial communities and soil organic matter (SOM) biogeochemistry through a variety of mechanisms. To investigate the impact of chronic warming, N-addition, and simultaneous warming and N-addition (warming + N) on forest soils, soil samples from the Harvard Forest Soil Warming and Nitrogen Addition (SWaN) experiment were analyzed after 14 years. Elemental analysis, targeted compound analysis by gas chromatography-mass spectrometry, and 13 C nuclear magnetic resonance (NMR) spectroscopy were used to analyze changes in SOM in both the organic and mineral (0-10 cm) soil layers. Overall, changes in the molecular composition and microbial biomass were observed, but the extent of differences was unique to warming, N-addition, and warming + N treatments. Specifically, Naddition slowed SOM decomposition as measured via solid-state 13 C NMR, while warming and warming + N accelerated SOM decomposition. Continued SOM decomposition after 14 years with warming + N signified a pronounced change to observations made after 4 and 10 years of experimental treatment. This is also demonstrative of how a two-factor approach leads to a unique molecular-level response that cannot be predicted from experiments with individual stressors alone. This study emphasizes the need to observe environmental stressors in tandem using a combination of molecular-level approaches to obtain a comprehensive understanding of how persistent anthropogenic activity will fundamentally alter forest soil systems.]]></ab></abstract>
		</profileDesc>
	</teiHeader>
	<text><body xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xlink="http://www.w3.org/1999/xlink">
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; INTRODUCTION</head><p>Forest soils represent one of Earth's greatest carbon (C) reservoirs <ref type="bibr">1,</ref><ref type="bibr">2</ref> and are a vital part of the global C cycle. <ref type="bibr">3</ref> Persistent anthropogenic activity can change both the biotic and abiotic character of these ecosystems, <ref type="bibr">4,</ref><ref type="bibr">5</ref> hence impacting the manner in which forests participate in global biogeochemical cycles. <ref type="bibr">[6]</ref><ref type="bibr">[7]</ref><ref type="bibr">[8]</ref> As atmospheric temperatures steadily rise globally, 9,10 continuous warming may limit capacity for soils to sequester C due to enhanced microbial respiration. <ref type="bibr">11,</ref><ref type="bibr">12</ref> Warming may also enhance both above-and below-ground C inputs to soil and contribute to the sensitive balance of soil C stored versus that respired. <ref type="bibr">[13]</ref><ref type="bibr">[14]</ref><ref type="bibr">[15]</ref><ref type="bibr">[16]</ref> In addition, increased atmospheric nitrogen (N) deposition to soil is currently occurring at a rate triple than that of 150 years ago <ref type="bibr">17</ref> and can increase the amount of C sequestered in forests. <ref type="bibr">[18]</ref><ref type="bibr">[19]</ref><ref type="bibr">[20]</ref> Moreover, warming and N-addition can impact the chemistry of soil organic matter (SOM), a complex mixture comprising plant, animal, and microbial matter at differing stages of decay, <ref type="bibr">21,</ref><ref type="bibr">22</ref> and soil microbial communities, <ref type="bibr">23,</ref><ref type="bibr">24</ref> which are both intrinsically linked to soil C and the global C cycle. <ref type="bibr">25</ref> Overall, the link between changing climate conditions and the expectation that persistent human activity will continue to accentuate stressors <ref type="bibr">26</ref> demonstrates the need for further information on the long-term impacts to SOM biogeochemistry.</p><p>Soil warming has contributed to the loss of soil C in forests <ref type="bibr">11,</ref><ref type="bibr">14,</ref><ref type="bibr">27,</ref><ref type="bibr">28</ref> and increased SOM decomposition. <ref type="bibr">14,</ref><ref type="bibr">[29]</ref><ref type="bibr">[30]</ref><ref type="bibr">[31]</ref><ref type="bibr">[32]</ref> These impacts have been attributed to changes in microbial biomass and activity <ref type="bibr">31,</ref><ref type="bibr">33,</ref><ref type="bibr">34</ref> and/or their substrate utilization. <ref type="bibr">[35]</ref><ref type="bibr">[36]</ref><ref type="bibr">[37]</ref> Enhanced decomposition of preferred microbial substrates, such as sugars and aliphatic lipids, was noted with short-term warming. <ref type="bibr">30,</ref><ref type="bibr">31,</ref><ref type="bibr">38</ref> Additionally, Feng et al. <ref type="bibr">30</ref> observed enhanced processing of longer-lived SOM compounds, such as lignin, due to increased fungal biomass with short-term warming. In contrast, long-term warming experiments have demonstrated variability in the magnitude of SOM decomposition, as evidenced by changes in microbial biomass and community structure <ref type="bibr">[39]</ref><ref type="bibr">[40]</ref><ref type="bibr">[41]</ref><ref type="bibr">[42]</ref> which altered the concentrations of both short-and long-lived SOM compounds. <ref type="bibr">34</ref> Melillo et al. <ref type="bibr">15</ref> hypothesized that microbial communities go through multiple "phases" of reorganization with long-term warming, being associated with variation in soil C respired over time. Interestingly, chronic warming was reported to alter the molecular-level composition of SOM even though microbial biomass decreased. <ref type="bibr">34</ref> Observed changes in the SOM chemical composition reflected a decline in polysaccharides, a preferred microbial substrate, and was likely due to a more active, yet smaller, microbial community. <ref type="bibr">34,</ref><ref type="bibr">39</ref> Furthermore, Pec et al. <ref type="bibr">43</ref> reported that short-versus long-term warming altered SOM composition uniquely, in addition to impacts to the soil fungal community composition. Similarly, detailed molecular-level characterization of SOM with warming demonstrated that the extent of changes to specific SOM components intensified with time <ref type="bibr">31,</ref><ref type="bibr">32</ref> emphasizing the need to better elucidate how warming alters SOM in the long term. Model predictions suggest that warming will continue to enhance soil C losses, <ref type="bibr">12</ref> which further demonstrates the need to better understand the magnitude with which warming will enhance SOM degradation and alter SOM chemistry over time. <ref type="bibr">31,</ref><ref type="bibr">32</ref> In contrast to warming, N-additions to temperate deciduous forest soils tend to increase C sequestration, <ref type="bibr">18,</ref><ref type="bibr">[44]</ref><ref type="bibr">[45]</ref><ref type="bibr">[46]</ref><ref type="bibr">[47]</ref><ref type="bibr">[48]</ref> lower microbial biomass and diversity, <ref type="bibr">20,</ref><ref type="bibr">49,</ref><ref type="bibr">50</ref> and decrease SOM decomposition. <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">51</ref> Both short-and long-term experiments have demonstrated that the addition of N increases C storage through slowed or suppressed microbial SOM decomposition. <ref type="bibr">18,</ref><ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">51,</ref><ref type="bibr">52</ref> Primarily, altered biogeochemical processing of plant-derived compounds is observed <ref type="bibr">51</ref> and generally attributed to changes in microbial biomass, community composition, and enzymatic activity. <ref type="bibr">18,</ref><ref type="bibr">50,</ref><ref type="bibr">53</ref> For instance, Hasegawa et al. <ref type="bibr">54</ref> demonstrated that the decomposition of lignin was suppressed after 12 years of N-addition, an observation commonly linked to decreased fungal biomass and activity, <ref type="bibr">18,</ref><ref type="bibr">48,</ref><ref type="bibr">55,</ref><ref type="bibr">56</ref> a microbial group imperative in the processing of lignin. <ref type="bibr">57</ref> Molecular-level studies also reported decreased SOM decomposition after 4 <ref type="bibr">31</ref> and 10 years 32 of Naddition, likely due to the accumulation of plant-derived inputs and also suppressed microbial decomposition of SOM. This suggests that while N-addition will continue to suppress SOM decomposition in forest soils, it may do so through a variety of distinct mechanisms, including via degradation by enzymes, <ref type="bibr">58</ref> suppressed microbial activity, <ref type="bibr">18</ref> and altered substrate utilization, <ref type="bibr">59</ref> which together led to a variable response over time.</p><p>Warming and N-addition are both environmental stressors that have been extensively studied independently; however, fewer studies have investigated the impact of warming and Naddition when combined (warming + N). <ref type="bibr">[60]</ref><ref type="bibr">[61]</ref><ref type="bibr">[62]</ref><ref type="bibr">[63]</ref><ref type="bibr">[64]</ref> Crucially, Rillig et al. <ref type="bibr">65</ref> indicated that the study of singular environmental stressors are not appropriate models for complex systems. This is especially relevant for forests, considering how warming and N-addition antagonistically alter soil C, <ref type="bibr">28,</ref><ref type="bibr">45</ref> microbial communities, <ref type="bibr">18,</ref><ref type="bibr">20,</ref><ref type="bibr">34</ref> and specific SOM compound concentrations, <ref type="bibr">30,</ref><ref type="bibr">44</ref> as well as how these responses are individually modulated by time. <ref type="bibr">43,</ref><ref type="bibr">51</ref> Previous studies focused on how warming + N will alter soil biogeochemistry noted that while the relationship between warming and N-addition is often in opposition, specific molecular responses vary with time. <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">62,</ref><ref type="bibr">66</ref> For example, soil C did not significantly change with warming + N, <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">66</ref> and microbial biomass either increased <ref type="bibr">66</ref> or did not change. <ref type="bibr">62</ref> Furthermore, studies have demonstrated that in the short term, warming + N alters SOM molecular composition more similar to N-addition alone. <ref type="bibr">31</ref> However, after 10 years, vandenEnden et al. <ref type="bibr">32</ref> reported that warming + N combined no longer mirrored the impact to SOM composition as N-addition alone and was more similar to warming alone. A complex web of interactions between microbial communities, C pools, and organic matter chemistry, <ref type="bibr">21</ref> as well as the difficulty of predicting responses over time, ultimately delineates the need for further research into how warming + N will alter SOM. Moreover, given that warming and N-addition alter soil C biogeochemistry distinctly, predicting how forest soils will respond in the long term requires further study, especially using molecular-level SOM characterization techniques.</p><p>Consequently, our aim in this study was to identify how warming, N-addition, and warming + N alter SOM chemistry and cycling following 14 years of experimental treatment. Soil samples were collected from the Soil Warming and Nitrogen Addition (SWaN) Study at the Harvard Forest Long-term Ecological Research (LTER) site and assessed using a combination of targeted and non-targeted molecular-level techniques. We used two complementary analytical approaches including gas chromatography-mass spectrometry and solidstate <ref type="bibr">13</ref> C nuclear magnetic resonance (NMR) spectroscopy to characterize changes in SOM chemistry. <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">67</ref> When used in tandem, these methods quantify distinct classes of plant-and microbial-derived compounds that is combined with an overview of SOM chemistry from solid-state <ref type="bibr">13</ref> C NMR spectroscopy. We also performed C and N analyses, which are indicative of general pool sizes and offer insights into substrate availability. We hypothesized that warming alone would continue to accelerate SOM decomposition via preferential use of specific SOM compounds, while N-addition would continue to suppress SOM degradation and result in the preservation of SOM compounds. We also hypothesized that warming + N would mirror warming and continue to accelerate SOM decomposition, but not to the same extent as warming alone <ref type="bibr">31,</ref><ref type="bibr">32</ref> given that warming impacts can be variable over time. <ref type="bibr">42</ref> Overall, this study investigates how SOM chemistry is altered after 14 years of warming, N-addition, and warming + N, with the goal of obtaining molecular-level insights into how different anthropogenic stressors, alone and combined, uniquely impact forest soils.</p><p>&#9632; METHODS Site Experimental Design and Sample Collection. Soil samples were collected from the SWaN plots at Harvard Forest located in Petersham, Massachusetts, USA, a forest characterized by a closed canopy made up primarily of hardwoods including red oak, red maple, and American beech. <ref type="bibr">66</ref> This forest has an average mean temperature of 7 &#176;C and a mean annual precipitation of 110 cm. <ref type="bibr">68</ref> Harvard forest soils are characterized as Gloucester-series with a fine loam texture. <ref type="bibr">66,</ref><ref type="bibr">69</ref> The SWaN plots (3 m &#215; 3 m) include the following treatments: ambient (control), warming, nitrogen addition (N-addition), or simultaneous warming + N addition (warming + N), and each treatment is replicated six times. Warming and warming + N plots are heated with buried cables running 10 cm below the soil surface, continuously elevating the temperature to 5 &#176;C above ambient soil temperature. Nitrogen is supplied to N-addition and warming + N plots monthly from May through October via addition of aqueous ammonium nitrate at a rate equivalent to 5 g N m -2 y -1 . Further information about the SWaN experimental design may be found in previously published studies. <ref type="bibr">18,</ref><ref type="bibr">66,</ref><ref type="bibr">69</ref> All plots have similar above-ground inputs as there are no canopy trees within the plots. Soil samples were collected in July of 2020 (after 14 years of treatment). For each plot, two individual samples were taken and included a 10 cm &#215; 10 cm forest floor (organic horizon) sample followed by 0-10 cm core (5 cm diameter) of mineral soil. Soils were stored at 4 &#176;C and then passed through a 2 mm sieve to remove rocks, roots, and organic debris &gt;2 mm. Replicate cores were pooled and homogenized, and then, soil moisture was determined gravimetrically by drying organic horizon and mineral soil subsamples at 60 and 105 &#176;C, respectively. Additional fresh subsamples were immediately frozen at -20 &#176;C and then freeze dried and finely ground for subsequent molecular analyses. <ref type="bibr">70</ref> Soil Carbon and Nitrogen Measurements. Total organic C and N were measured using a Thermo Flash 2000 Elemental Analyzer which treated samples with O 2 gas at a temperature of 950 &#176;C. Each soil sample was measured in duplicate.</p><p>Targeted Analysis of SOM Compounds. A sequential extraction method which targets various plant-and microbialderived compounds was used. <ref type="bibr">71</ref> Each soil sample was extracted in duplicate. First, all solvent-extractable sugars and lipids were isolated from samples. <ref type="bibr">72</ref> Approximately 1 g of soil was used for organic horizon samples, while 2 g of soil was used for mineral (0-10 cm) layer samples. Samples were extracted via sonification with 30 mL of dichloromethane (Optima grade, Fisher Scientific), 1:1 dichloromethane/methanol and methanol (Optima grade, Fisher Scientific), respectively, and then centrifuged at 2500 rpm for 5 min. Samples were next filtered using two glass microfiber filter papers (GF/A and GF/F, Whatman), both with a diameter of 55 mm and a pore size of 0.7 and 1.6 &#956;m, respectively. Following filtration, extracts were concentrated using rotary evaporation. Samples were placed into 2 mL glass vials and evaporated to dryness using N 2 gas prior to quantification using gas chromatography-mass spectrometry.</p><p>Base hydrolysis was used to cleave ester-bound lipids in cutin-, suberin-, and microbial-derived compounds. <ref type="bibr">73</ref> Soil samples that were previously solvent-extracted were placed within Teflon-lined bombs with 15 mL of 1 M methanolic (Optima grade, Fisher Scientific) potassium hydroxide (ACS grade, Fisher Scientific) and heated for 3 h at a constant temperature of 100 &#176;C. Following heating, samples were cooled and extracted with 1:1 dichloromethane/methanol (Optima grade, Fisher Scientific) and centrifuged twice (10 min at 2500 rpm). The supernatant was isolated and acidified using 6 M hydrochloric acid (ACS grade, Sigma-Aldrich) to a pH of 1. Following concentration via rotary evaporation, liquid-liquid extraction was performed using anhydrous ethyl ether (ACS grade, Fisher Scientific) and deionized water (18.2 M&#937; cm). Samples were then dried over anhydrous sodium sulfate (ACS grade, Fisher Scientific), concentrated using rotary evaporation, and dried using N 2 gas prior to quantification using gas chromatography-mass spectrometry.</p><p>Copper(II) oxide (&gt;99%, Fisher Scientific) was used to isolate vanillyl, syringyl, and cinnamyl phenols from lignin. <ref type="bibr">74</ref> Previously extracted soil samples were placed within Teflonlined bombs with 1 g of copper(II) oxide (&gt;99%, Fisher Scientific), 100 mg of ammonium iron(II) sulfate hexahydrate (ACS grade, Sigma-Aldrich), and 15 mL of 2 M sodium hydroxide (ACS grade, Fisher Scientific) in deionized water (18.2 M&#937; cm). Samples were purged using N 2 gas prior to being heated at 170 &#176;C for 2.5 h. Samples were cooled, deionized water (18.2 M&#937; cm) was added, and the mixture was centrifuged (8 min at 2500 rpm). Samples were maintained in the dark at room temperature for 1 h to prevent polymerization of cinnamic acids. <ref type="bibr">75</ref> Lignin-derived compounds were isolated using solid-phase extraction <ref type="bibr">76</ref> with 60 mg of Oasis HLB cartridges (Waters Corporation). Solid-phase extraction cartridges were conditioned using methanol (Optima grade, Fisher Scientific) and deionized water (18.2 M&#937; cm) prior to the sample loading and washing using 30% methanol (Optima grade, Fisher Scientific) in deionized water (18.2 M&#937; cm). Samples were eluted using a mix of 70:25:5 dichloromethane (Optima grade, Fisher Scientific): methyl acetate (&gt;99.8%, Sigma-Aldrich): pyridine (ACS grade, Fisher Scientific) and methanol (Optima grade, Fisher Scientific), dried over anhydrous sodium sulfate (ACS grade, Fisher Scientific), and concentrated in 2 mL glass vials using N 2 gas.</p><p>Phospholipid fatty acids (PLFAs) were extracted to measure microbial biomass and community composition. <ref type="bibr">70</ref> Approximately 1 and 5 g of soil were used for organic and mineral samples, respectively. Samples were mixed with 6.3 mL of a 0.15 M citrate buffer, 15.8 mL of methanol (Optima grade, Fisher Scientific), and 7.9 mL of chloroform (Optima grade, Fisher Scientific) and shaken for 24 h. Afterward, samples were centrifuged at 2500 rpm for 10 min and separated into fractions following the addition of chloroform (Optima grade, Fisher Scientific) and citrate buffer. The chloroform (Optima grade, Fisher Scientific) phase was dried under N 2 gas. Extracts were redissolved in 1 mL of chloroform (Optima grade, Fisher Scientific), and compounds were separated through silicic acid columns. Following fractionation of the extract, the fraction containing polar lipids was treated with methanolic (Optima grade, Fisher Scientific) potassium hydroxide (ACS grade, Fisher Scientific) and heated, converting these compounds to fatty acid methyl esters. These compounds were further extracted using a 4:1 mixture of hexanes (Optima grade, Fisher Scientific) and chloroform (Optima grade, Fisher Scientific) <ref type="bibr">77</ref> and transferred into 2 mL glass vials prior to analysis via gas chromatography-mass spectrometry.</p><p>Targeted SOM Compound Quantification and Data Analysis. Dried isolated extracts were stored at -15 &#176;C prior to quantification using gas chromatography-mass spectrometry. Prior to analysis, samples were derivatized using N,Obistrifluoroacetamide (&gt;98.5%, Sigma-Aldrich) and anhydrous pyridine (&gt;99.8%, Sigma-Aldrich). Base hydrolysis extracts were further derivatized using N,N-dimethylformamide dimethyl acetal (ACS grade, Fisher Scientific) to methylate compounds. <ref type="bibr">78</ref> External standards were prepared and derivatized using the same technique/materials. The external standards included 1-docosanol (&gt;98%, Sigma-Aldrich), methyl tricosanoate (&gt;99%, Sigma-Aldrich), cholesterol (&gt;99%, Sigma-Aldrich), and tetracosane (&gt;99%, Sigma-Aldrich) and were used for solvent extraction samples. Methyl tricosonoate (&gt;99%, Sigma-Aldrich) was used for base hydrolysis, and syringaldehyde (&gt;98%, Sigma-Aldrich) and syringic acid (&gt;98%, Sigma-Aldrich) were used for copper oxidation compounds. Oleic acid (&gt;99%, Sigma-Aldrich) was used as an external standard for PLFA quantification.</p><p>Samples and standards were analyzed using an Agilent 7890B gas chromatograph hyphenated to an extractor ion source operated in the electron impact ( <ref type="formula">70</ref> C NMR Spectroscopy. Composite samples were prepared for both organic and mineral layer samples for each treatment. This approach was used for consistency to previous studies that collected NMR data on these plots after 4 and 10 years of the experiment. <ref type="bibr">31,</ref><ref type="bibr">32</ref> For each treatment, 2 g of each mineral layer replicate was measured, ground, and combined to create a composite sample, whereas 0.1 g was used for each organic layer replicate. Mineral layer composites were repeatedly treated with 10% hydrofluoric acid (ACS grade, BDH Chemicals) to remove any paramagnetic materials, hence enhancing signal and improving the observed signal-to-noise ratio. <ref type="bibr">79</ref> Following repeated extraction with hydrofluoric acid, samples were washed with deionized water (18.2 M&#937; cm) until the electrical conductivity of the supernatant indicated low salt contents. Samples were then frozen and freeze-dried prior to analysis.</p><p>Samples were packed into 4 mm zirconium rotors, sealed with a Kel-F cap, and then placed within a 500 MHz Bruker BioSpin Avance III spectrometer with a 4 mm H-X MAS probe. Both organic and mineral composite samples were acquired using 21,000 scans under magic angle spinning (MAS) at a rate of 11 kHz. NMR data were acquired using a 1 ms ramp-cross-polarization contact time and with a 1 s recycle delay. <ref type="bibr">80</ref> NMR spectra were analyzed using Bruker TopSpin 3.6.2., and spectra were divided into four chemical shift regions: alkyl C (0-50 ppm), O-alkyl C (50-110 ppm), aromatic and phenolic C (110-165 ppm), and carboxyl and carbonyl C (165-215 ppm). <ref type="bibr">81</ref> The integration of these four regions of the NMR spectra provides information regarding the amount of each compound class contained in each sample. The alkyl/O-alkyl C ratio was calculated by dividing the area of alkyl C by that of O-alkyl C, and this ratio is used to provide a relative indicator of SOM degradation. <ref type="bibr">67,</ref><ref type="bibr">81</ref> Studies have observed increasing alkyl/O-alkyl C ratios with progressive decomposition of SOM. <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">51</ref> We also compared the results from this study (14 years of experimental treatment) to results obtained after 4 and 10 years of the experiment. <ref type="bibr">31,</ref><ref type="bibr">32</ref> Statistical Analyses. Soil C and N contents were analyzed for statistically significant differences (p &#8804; 0.05) using a oneway analysis of variance (ANOVA) and post-hoc Bonferroni analysis [SPSS Statistics (Software; version 28)], using field replication (n = 5) as a fixed factor. Differences in SOMtargeted compounds between treatment and control plots were tested for statistical significance (p &#8804; 0.05) using ANOVA) with post-hoc Bonferroni analysis (SPSS Statistics Software; version 28) with field replication constrained as a fixed factor (n = 5) and analytical replication set as a random factor (n = 2). We also compiled select SOM compound concentrations from years 4, 10, and 14 that represent a gradient of relative stability in soil for a broader analysis using non-metric multidimensional scaling (NMDS). Concentrations for aliphatic compounds, simple sugars, and cutin-, suberin-, and lignin-derived compounds were compiled and compared for both organic and mineral soil horizons. We transformed the  <ref type="table">S1</ref> and<ref type="table">S2</ref>).</p><p>data by taking the square root of the relative abundance of each compound within a plot to correct the skewness in the data set across the three sampling years and to place less importance on SOM compounds found in high concentrations. We tested for significance between groups using a PERManova (p &#8804; 0.05). We used the vegan package <ref type="bibr">82</ref> of R <ref type="bibr">83</ref> to perform all multivariate analyses.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; RESULTS</head><p>Soil Carbon and Nitrogen Contents. In the organic horizon, warming, N-addition, and warming + N plots, all had lower total soil C relative to the control plots (Figure <ref type="figure">S1</ref>), but these differences were not statistically significant. Similarly, in the mineral layer, no significant differences in soil C were observed (Figure <ref type="figure">S1</ref>), but decreases relative to the control plots were noted. No significant changes were observed with total soil N content in both layers (Figure <ref type="figure">S1</ref>).</p><p>Targeted Analysis of SOM Compounds. Solvent-Extractable Compounds. Solvent-extractable compounds from both organic and mineral samples included a variety of lipids arising from both plant and microbial sources (Tables <ref type="table">S1</ref> and<ref type="table">S2</ref>). Solvent-extractable compounds included aliphatic lipids (n-alkanes, n-alkanols, and n-alkanoic acids), cyclic lipids (terpenoids and plant-derived steroids), and sugars. The concentrations of both long-chain (&gt;C 20 ) aliphatic lipids (nalkanes, n-alkanols, and n-alkanoic acids) and short-chain (&lt;C 20 ) aliphatic lipids (n-alkanoic acids) provide information about the microbial-or plant-derived sources of SOM, <ref type="bibr">71</ref> and differences in the distribution of these compounds (Table <ref type="table">S1</ref>) indicated variable warming impacts to the SOM composition. n-Alkanes and n-alkanols with chain lengths less than C 20 were not detected (Tables <ref type="table">S1</ref> and<ref type="table">S2</ref>). Overall, we did not observe any significant differences in solvent extractable compound concentrations with warming, N-addition, and warming + N in the organic or mineral soil layers (Figure <ref type="figure">1</ref>; Tables <ref type="table">S1</ref> and<ref type="table">S2</ref>). Warming decreased the total concentration of aliphatic lipids within the organic horizon and increased the aliphatic lipids in the mineral soil (Figure <ref type="figure">1</ref>) but not significantly. In both the organic and mineral layer, N-addition increased the overall concentration of aliphatic lipids (both long-chain and shortchain compounds) but not significantly (Figure <ref type="figure">1</ref>; Table <ref type="table">S1</ref>). Warming + N decreased the concentration of aliphatic lipids (both long-chain and short-chain lipids in both soil layers) but not significantly (Figure <ref type="figure">1</ref>; Table <ref type="table">S1</ref>). Cyclic lipids were not significantly changed with any treatment or in any soil layer (Table <ref type="table">S2</ref>). Similarly, sugar concentrations were not significantly different with warming, N-addition, or warming + N in the organic and mineral layers (Table <ref type="table">S2</ref>). A  <ref type="table">S3</ref>).  <ref type="table">S4</ref>).</p><p>comparison of aliphatic lipid and sugar concentrations over time using NMDS (Figure <ref type="figure">5</ref>) shows more similarities between treatments at years 10 and 14 (p = 0.008) in comparison to year 4.</p><p>Cutin-and Suberin-Derived Compounds. Cutin-and suberin-derived compounds, which are derived from aboveand below-ground plant sources, respectively, <ref type="bibr">84</ref> were detected in both organic and mineral layer samples (Figure <ref type="figure">2</ref>; Table <ref type="table">S3</ref>) but to varying extents. In the organic horizon, there were no significant differences in cutin-or suberin-derived compounds across treatments. In the mineral soil, we observed significantly higher concentrations of compounds found in cutin and suberin with warming and warming + N. The ratio of &#969;hydroxyalkanoic acids to total hydrolyzable acids of corresponding chain length (&#969;-C 16 /&#931;C 16 ; &#969;-C 18 /&#931;C 18 ) <ref type="bibr">71,</ref><ref type="bibr">72</ref> has been used to assess the degree of degradation of cutin-derived  <ref type="table">S5</ref>). compounds. There was no significant difference in this ratio with warming, N-addition, or warming + N in both the organic and mineral soils relative to the control (Table <ref type="table">S3</ref>), indicating that cutin decomposition was not enhanced with treatment. Cutin-and suberin-derived compounds were included in the NMDS comparison and further demonstrate a change in biogeochemical trajectory over time (Figure <ref type="figure">5</ref>).</p><p>Microbial-Derived Lipids. Microbial-derived lipids, which included C 14 -C 19 branched alkanoic acids, long-chain nalkanols (C 18 ), and long-chain n-alkanoic acids (C 16 + C 18 ), <ref type="bibr">85</ref> were observed in both the organic and mineral layers (Figure <ref type="figure">3</ref>; Table <ref type="table">S4</ref>). These lipids encompass both cellular and extracellular lipids but differ from PLFAs as they are not membrane lipids. <ref type="bibr">86</ref> In the organic horizon, there was no significant variation with warming, N-addition, or warming + N. However, in the mineral soil, warming and warming + N significantly increased microbial-derived lipid concentrations.</p><p>Lignin-Derived Compounds. Lignin-derived compounds, which included vanillyl, syringyl, and cinnamyl phenols, were detected in both organic and mineral soil samples (Figure <ref type="figure">4</ref>; Table <ref type="table">S5</ref>). With warming, syringyl phenols were significantly higher in the organic horizon, but warming did not change the total lignin-derived compound concentrations significantly relative to the control in either soil layer (Figure <ref type="figure">4</ref>). Acid-toaldehyde ratios for vanillyl and syringyl compounds (Ad v /Al v ; Ad s /Al s ) are used to assess lignin oxidation. <ref type="bibr">74</ref> However, no significant changes were observed in either the organic or mineral soil with any treatment (Table <ref type="table">S5</ref>). In comparison to other classes of SOM compounds by NMDS (Figure <ref type="figure">5</ref>), lignin phenol concentrations were more like year 10 than year 4.</p><p>Phospholipid Fatty Acids. PLFA concentrations were altered with warming, N-addition, and warming + N treatments (Figure <ref type="figure">6</ref>; Table <ref type="table">S6</ref>). N-addition and warming + N exhibited significantly lower microbial biomass (total PLFAs) in the organic horizon. There were no significant changes to total microbial biomass in the mineral soil. Differences in the soil microbial community composition were also observed. With warming, the ratio of Gram-negative to Gram-positive bacteria was significantly lower in the organic horizon. However, the microbial community composition did not change significantly in mineral soil samples (Table <ref type="table">S6</ref>). Naddition did not lead to any significant change with regard to community composition in either the organic or mineral layers (Table <ref type="table">S6</ref>). Warming + N significantly lowered the ratio of Gram-negative to Gram-positive bacteria but did not significantly change the community composition in the mineral layer (Table <ref type="table">S6</ref>). Specific ratios of PLFAs (monoenoic: saturated PLFAs, cy17:0/16:1&#969;7c, cy19:0/18:1&#969;7c) have been reported to increase when microbial biomass is faced with substrate limitations or other stress, such as changes in the size of the biomass and shifts in the community composition. <ref type="bibr">87</ref> In the organic horizon, warming and warming + N decreased the ratio of monoenoic to saturated PLFAs significantly relative to the control. Furthermore, the ratio of cy17:0/16:1&#969;7c increased significantly with warming, and the ratio of cy19:0/ 18:1&#969;7c increased significantly with warming + N (Table <ref type="table">S6</ref>). No significant changes in any stress ratios were noted with any treatment in the mineral soil (Table <ref type="table">S6</ref>).</p><p>Solid-State 13 C NMR Spectroscopy. Using 13 C NMR spectroscopy, changes in the integration values of chemical shift regions representing different forms of SOM relative to the control were observed. The integrated regions (Table <ref type="table">S7</ref>) included alkyl C region (0-50 ppm), which represents resonances from aliphatic lipids and side chains from cutin and suberin; O-alkyl C region (50-110 ppm), which arises from peptides and carbohydrates, such as cellulose and methoxy C from lignin; aromatic and phenolic C region (110-165 ppm), which includes lignin and aromatic amino acids, and the carboxyl and carbonyl C region (165-215 ppm), which stems from a range of side-chain functionalities from lipids, peptides, and other oxidation products. <ref type="bibr">88</ref> Warming increased alkyl C and decreased aromatic and phenolic C relative to the control in the organic horizon but did not impact O-alkyl C or aromatic and phenolic C (Table <ref type="table">S7</ref>). In the mineral soil, aromatic and phenolic C, as well as carboxyl and carbonyl C, differed from the control (Table <ref type="table">S7</ref>). Naddition altered alkyl, O-alkyl, and carboxyl and carbonyl C in both organic and mineral layers, respectively, increasing (Oalkyl C, carboxyl, and carbonyl C) and decreasing (alkyl C) these integration values relative to the control (Table <ref type="table">S7</ref>). Warming + N increased alkyl C and decreased carboxyl and carbonyl in both organic and mineral layers relative to the control (Table <ref type="table">S7</ref>). Furthermore, warming + N decreased Oalkyl C and aromatic and phenolic C in the organic layer only (Table <ref type="table">S7</ref>). The alkyl/O-alkyl C ratio is used to assess the degree of SOM degradation relative to the control. <ref type="bibr">81</ref> Warming enhanced SOM degradation in the organic horizon but did not change the relative state of SOM decomposition in the mineral layer. With N-addition, suppressed SOM degradation was observed in both the organic and mineral layers. Warming + N accelerated SOM degradation in both the organic and mineral soils relative to the control (Figure <ref type="figure">7</ref>). The relative change in SOM decomposition over time shows that some responses have intensified depending on the horizon, whereas others have weakened (Figure <ref type="figure">7</ref>).</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; DISCUSSION</head><p>Overall, we observed differences in the SOM chemical composition after 14 years of warming, N-addition, and warming + N. However, these changes were unique to each treatment and reflect different patterns and perturbations to biogeochemical cycling of soil C. Using several molecular proxies, we observed that warming and warming + N accelerated SOM degradation, while N-addition suppressed the decomposition of SOM. Furthermore, the magnitude and direction of these differences varied, especially when comparing warming alone to warming + N. Warming + N demonstrated the greatest enhancement in SOM decomposition, indicating that the combination of two antagonistic stressors (warming vs N-addition) results in unique changes to SOM biogeochemistry. Below, we outline in detail how after 14 years, distinct shifts in SOM chemistry were observed with warming, N-addition, and warming + N. We also compare these results to our earlier studies at this same site, which includes the same characterization approach after 4 years, published in a study by Pisani et al. <ref type="bibr">31</ref> and 10 years, published in a study by vandenEnden et al. <ref type="bibr">32</ref> of the experiment (Figures <ref type="figure">5</ref> and<ref type="figure">7</ref>).</p><p>Soil Warming. Several studies showed that continued warming accelerates SOM decomposition. <ref type="bibr">14,</ref><ref type="bibr">29,</ref><ref type="bibr">30</ref> Some studies have indicated that the flux of soil CO 2 from microbial respiration with prolonged warming is variable and dependent on the use of substrates by soil microbes. <ref type="bibr">89,</ref><ref type="bibr">90</ref> Furthermore, the magnitude of warming impacts is also dependent on time, and observed changes in the SOM composition, soil respiration, and overall microbial activity have been reported to respond non-linearly with continued warming. <ref type="bibr">31,</ref><ref type="bibr">32,</ref><ref type="bibr">39,</ref><ref type="bibr">40</ref> Other studies have indicated that above-and below-ground inputs can increase with soil warming. <ref type="bibr">13,</ref><ref type="bibr">30</ref> In this study, the organic horizon exhibited decreased soil C concentrations (Figure <ref type="figure">S1</ref>) from 28 to 22%, although not significantly, which suggests that prolonged warming likely continued to enhance microbial decomposition of soil C. <ref type="bibr">91</ref> Similarly, solid-state <ref type="bibr">13</ref> C NMR spectroscopy indicated an overall increase in SOM degradation based on the alkyl/O-alkyl C ratio. <ref type="bibr">81</ref> Microbial biomass (PLFA concentrations) did not change significantly but differences in microbial community composition observed were consistent with possible changes in substrate use (Table <ref type="table">S6</ref>). After 10 years of warming, vandenEnden et al. <ref type="bibr">32</ref> also reported no change in total soil microbial biomass indicating that the observed enhanced SOM decomposition was likely due to shifts in microbial substrate utilization. Frey et al. <ref type="bibr">41</ref> reported that long-term warming altered substrate utilization patterns in an adjacent soil warming experiment at Harvard Forest (Figure <ref type="figure">S1</ref> and Table <ref type="table">S6</ref>). Anthony et al. <ref type="bibr">92</ref> reported that fungal community evenness declined, and the community composition was altered with 10 years of soil warming which could also impact substrate utilization patterns. We observed that the ratio of Gram-negative to Gram-positive bacteria shifted, suggesting that bacteria may be responding to changes in substrate availability. <ref type="bibr">93</ref> Furthermore, we observed increased production of cyclic PLFAs (cy17:0/16:1&#969;7c and monoenoic/ saturated ratios; Table <ref type="table">S6</ref>) which signify increased stress with warming, likely related to changes in substrate availability or continuous soil warming, which has been shown to increase microbial stress independent of substrate availability. <ref type="bibr">94</ref> Interestingly, these shifts in microbial community structure did not reflect extensive changes in the use of specific SOM compounds, as previously observed in the organic horizon after 10 years of the experiment. <ref type="bibr">32</ref> After 14 years of the experiment, specific SOM compounds (Tables <ref type="table">S1-S5</ref>) did not exhibit the same extent of significant differences relative to the control organic horizon. However, enhanced SOM decomposition observed via solid-state <ref type="bibr">13</ref> C NMR spectroscopy (Figure <ref type="figure">7</ref>) indicated that changes to the microbial community composition continued to subsequently alter the SOM composition but to a lesser extent. <ref type="bibr">32</ref> Collectively, these results indicate that the extent to which warming has impacted SOM degradation has  <ref type="table">S7</ref>).</p><p>weakened over time in the forest floor and is consistent with the patterns observed for long-term warming. <ref type="bibr">43</ref> In contrast, the mineral (0-10 cm) soil did not exhibit any differences in total soil organic C concentrations or microbial biomass (PLFAs) and community composition (Figures <ref type="figure">S1</ref> &amp;<ref type="figure">6</ref>; Table <ref type="table">S6</ref>). Furthermore, short-chain lipids, cyclic lipids, and simple sugars (Tables <ref type="table">S1</ref> and<ref type="table">S2</ref>), which are typically categorized as preferred microbial substrates, were not statistically different after 14 years of warming. Higher concentrations of suberin-derived compounds relative to the control were observed, signifying continued preservation of these root-derived compounds with warming and/or increased below-ground inputs. <ref type="bibr">32</ref> There were also higher concentrations of C 16 , C 18 , and mid-chain lipids relative to the control in the mineral soil (Table <ref type="table">S3</ref>). Increased root-derived compounds are consistent with the results of Kwatcho Kengdo et al. <ref type="bibr">13</ref> who reported increased fine root turnover with 15 years of warming in a European mountain forest. Microbial-derived lipids, which are thought to result from plant-derived SOM processing and associated SOM turnover, <ref type="bibr">85</ref> increased in the mineral soil further suggesting enhanced decomposition of plant-derived compounds. Although some changes in specific SOM compounds were detected, there was no overall difference in the SOM composition observed by solid-state <ref type="bibr">13</ref> C NMR in the mineral layer (Figure <ref type="figure">S1</ref>) and no difference in SOM decomposition relative to the control (Figure <ref type="figure">7</ref>; Table <ref type="table">S7</ref>). These observations are distinct from previous studies where enhanced SOM decomposition was observed with continued warming <ref type="bibr">31,</ref><ref type="bibr">32</ref> but likely indicate that the overall composition of SOM reflects an apparent equilibrium between increased plantderived inputs and microbial decomposition.</p><p>After 10 years of warming, vandenEnden et al. <ref type="bibr">32</ref> reported a significant decrease in plant-derived aliphatic and cyclic lipids (steroids and terpenoids). In our study, no significant decrease was observed in either the organic or mineral soils after 14 years (Figure <ref type="figure">1</ref>). Furthermore, vandenEnden et al. <ref type="bibr">32</ref> reported that warming had elevated SOM degradation by 16 and 26% relative to the control in organic and mineral soils, respectively. After 14 years, we noted that warming had only elevated degradation by 6.5% in the organic horizon and not at all in mineral soil (0%) (Figure <ref type="figure">7</ref>). Our results are consistent with our hypothesis, which stipulates that warming will continue to accelerate SOM degradation. Furthermore, NMDS analysis shows similarities between concentrations of aliphatic compounds between years 10 and 14 but not year 4 (Figure <ref type="figure">5</ref>). However, the magnitude of accelerated SOM decomposition is very different from that observed at 4 and 10 years, <ref type="bibr">31,</ref><ref type="bibr">32</ref> demonstrating that sustained warming is dynamic and selectively impacts SOM compounds uniquely over time. N-Addition. Previous studies have indicated that Naddition slows SOM decomposition in temperate deciduous forests <ref type="bibr">45,</ref><ref type="bibr">54</ref> via the suppression of microbial processing of SOM. <ref type="bibr">51</ref> In our study, N-addition did not significantly alter soil C or N concentrations in both organic and mineral layers (Figure <ref type="figure">S1</ref>), which may be consistent with the time required to increase the soil C content with long-term N-addition. <ref type="bibr">18,</ref><ref type="bibr">95</ref> Furthermore, we observed a decrease in the alkyl/O-alkyl C ratio in both organic and mineral layers, suggesting that SOM decomposition was suppressed <ref type="bibr">81</ref> relative to the control (Figure <ref type="figure">7</ref>; Table <ref type="table">S7</ref>) and was consistent with observations made at years 4 and 10. In addition, microbial biomass was significantly lower with N-addition in the organic horizon (Figure <ref type="figure">6</ref>; Table <ref type="table">S6</ref>). Previous studies reported decreased microbial biomass with continued N-addition, related to the inhibition of oxidative enzyme activity, especially peroxidases which are responsible for the degradation of lignin. <ref type="bibr">18,</ref><ref type="bibr">59</ref> In the mineral layer, no significant changes were noted with regard to microbial biomass or community composition; however, the response of microbial communities and substrate utilization patterns may differ with soil depth. <ref type="bibr">46,</ref><ref type="bibr">96</ref> Furthermore, higher concentrations of long-chain aliphatic and cyclic (steroids and terpenoids) lipids were found in both soil layers (Figure <ref type="figure">1</ref>; Table <ref type="table">S1</ref> and<ref type="table">S2</ref>), which is indicative of suppressed processing of plant-derived inputs and SOM. <ref type="bibr">51</ref> An increase in sugars and O-alkyl C relative to the control further corroborates the slowing of SOM decomposition, given these compounds are preferred microbial substrates (Table <ref type="table">S7</ref>). <ref type="bibr">97</ref> No significant change was observed in the concentrations of microbialderived lipids (Table <ref type="table">S4</ref>) relative to the control in either soil layer which is consistent with previous studies that reported decreases in microbial biomass with N-addition. <ref type="bibr">51</ref> As such, the reduced microbial biomass and accumulation of plant-derived compounds with N-addition have subsequently reduced the production of cellular and extracellular microbial-derived lipids, suggesting that this group of compounds may be useful indicators to assess how anthropogenic change is altering soil C biogeochemical processes. Overall, the continued suppression of SOM decomposition in both horizons is consistent with previous observations as well as our hypothesis which stipulates that 14 years of continuous N-addition would suppress microbial biomass and SOM degradation in these temperate forest soils.</p><p>Soil Warming + N-Addition. When applied separately, warming and N-addition impact SOM differently. <ref type="bibr">51,</ref><ref type="bibr">61</ref> Therefore, it is important to study combined anthropogenic stressors to better understand how different components of global environmental change will alter soil C biogeochemistry in forests. Our previous investigations after 4 and 10 years documented <ref type="bibr">31,</ref><ref type="bibr">32</ref> that warming + N impacted SOM decomposition differently over time. With 4 years of experimental treatment, warming + N-addition behaved more similar to Naddition <ref type="bibr">31</ref> but after 10 years, warming + N accelerated SOM decomposition and was more similar to warming alone 32 (Figure <ref type="figure">7</ref>). After 14 years, and consistent with our hypothesis, warming + N has continued to mimic warming alone but to a different extent than observed previously. The soil C content was not significantly different to the control, but this does not reflect changes that have occurred with regard to specific SOM compounds or the overall rate of degradation. <ref type="bibr">98</ref> For example, alkyl/O-alkyl C ratios demonstrated that warming + N continued to accelerate degradation in both the organic and mineral layers through the microbial processing of preferred substrates (Figure <ref type="figure">7</ref>; Table <ref type="table">S7</ref>), despite a lack of increase in the microbial biomass. Notably, after 10 years, the rate of SOM degradation was accelerated with warming + N but not to the same extent as with warming alone (Figure <ref type="figure">7</ref>). <ref type="bibr">32</ref> Analysis of select SOM compounds by NMDS further confirms that warming + N results are more similar between years 10 and 14 as compared to year 4 (Figure <ref type="figure">5</ref>). Our results here demonstrate a shift in the biogeochemical trajectory with warming + N over time, however, exemplifying that the overall differences in the SOM composition are indeed distinct from warming alone. After 14 years, warming alone is no longer accelerating degradation, but warming + N is continuing to accelerate SOM decomposition in both soil layers. With warming + N, we also observed significantly lower microbial biomass (PLFA concentrations) in the organic horizon but no difference in microbial biomass in the mineral layer, indicating unique responses that are horizon-dependent (Figure <ref type="figure">6</ref>; Table <ref type="table">S6</ref>). In the organic horizon, increased microbial stress (cy19:0/ 18:1&#969;7c ratio) was observed with warming + N as compared to the other treatments which may be related to many factors including substrate stress and changes in the microbial environment. <ref type="bibr">99</ref> Interestingly, microbial stress (as measured via different PLFA ratios) has also been linked to temperature variance alone and has been shown to increase with higher temperatures independent of changes in substrate availability. <ref type="bibr">94</ref> This may explain why increased microbial stress was observed with both warming and warming + N but not with Naddition. Furthermore, Melillo et al. <ref type="bibr">15</ref> suggested that microbial community reorganization is indicative of long-term soil warming and respiration patterns, suggesting that these processes are dynamic over time. Our results indicate that as with warming, warming + N is also dynamic, and the changes in SOM decomposition are offset by increased plant inputs. As with warming, the concentrations of several specific SOM compounds did not change significantly (Figure <ref type="figure">1</ref>; Tables <ref type="table">S1</ref> and<ref type="table">S2</ref>), suggesting that the impact of warming + N has also weakened over time but not to the same magnitude as warming alone. More long-lived compounds, such as cutinand suberin-derived lipids, increased significantly in the mineral layer with both warming and warming + N (Figure <ref type="figure">2</ref>; Table <ref type="table">S3</ref>) and likely reflect the use of other preferred substrates as well as preservation of below-ground inputs that may also have shifted after 14 years of the experiment. <ref type="bibr">13</ref> Furthermore, there were higher concentrations of C 16 , C 18 and mid-chain lipids relative to the control in the mineral soil. As with warming, this may be due to a shift in microbial processing. This significant increase in microbial-derived lipids was observed with warming + N in the mineral soil (Figure <ref type="figure">3</ref>; Table <ref type="table">S4</ref>), indicating that despite no change in the size of the microbial biomass, the sustained metabolic activity continued to alter SOM biogeochemical trajectories and enhanced the overall SOM decomposition uniquely. Moreover, warming + N appears to have a more pronounced and long-lived impact on the SOM composition than warming alone after 14 years of the experiment, which is consistent with our third hypothesis.</p><p>As warming and N-addition alter SOM chemistry in opposite manners, <ref type="bibr">69</ref> it is difficult to predict how this combined treatment will continue to impact SOM decomposition. However, our hypothesis that warming + N will continue to mirror warming in both organic and mineral layers is supported by both NMR and molecular biogeochemistry data. Compared to previous studies, there is a smaller difference with warming + N relative to the control in both soil layers, consistent with CO 2 flux and expected warming behavior. Furthermore, we did not observe additive impacts with regard to warming and N-addition as other studies have noted. <ref type="bibr">100</ref> This may demonstrate that N-addition acts to offset certain warming-induced SOM responses and that these two environmental stressors do not work through the same mechanisms, as has been reported in other studies studying the response of SOM to warming + N. <ref type="bibr">101,</ref><ref type="bibr">102</ref> For instance, Song et al. <ref type="bibr">103</ref> proposed that N-addition may create a negative feedback loop and lessen the impact warming would otherwise have on SOM. However, most experiments are short term in nature and may not be sufficient in predicting how SOM will react after 14 years of continued treatment. In some regards, warming + N leads to a more pronounced change in SOM than warming alone, especially in the mineral layer, highlighting the need for further long-term studies of combined ecosystem stressors.</p><p>&#9632; CONCLUSIONS Following 14 years of treatment and using a combination of targeted and non-targeted techniques, we observed several changes to SOM degradation patterns with warming, Naddition, and warming + N in both organic and mineral (0-10 cm) soils. This approach offers insight into both specific compound concentrations associated with plant-and microbial-processing, as well as a holistic overview of the SOM molecular composition via solid-state <ref type="bibr">13</ref> C NMR spectroscopy. Warming and warming + N accelerated SOM degradation, whereas N-addition suppressed the decomposition of SOM, consistent with our initial hypotheses, as well as SOM behavior following 10 years of treatment. However, changes in the magnitude of SOM acceleration and suppression were noted, especially in the case of warming, wherein fewer molecular compound changes occurred in both organic and mineral soils. Likewise, while suppression of SOM degradation due to Naddition has been reported previously, 14 years of addition led to a stronger impact, especially in the mineral horizon. This is distinct from our previous observations after 4 and 10 years of experimental manipulations, wherein N-addition had a greater impact on the organic horizon. While warming + N continues to mirror warming alone, some aspects are markedly different, demonstrating the dynamic responses of SOM over time in forests. Overall, our study shows how two individually antagonistic, but very prevalent, environmental stressors interact with one another to change the character of SOM uniquely and over time. As persistent anthropogenic activity continues to heighten the concern surrounding these stressors, our results highlight the need for additional research studying the long-term and combined impact that warming + N, and ultimately, climate change will have upon forest soils. Our study further emphasizes that when combined, environmental stressors behave uniquely at the molecular-level, and these responses cannot be predicted from studies that consider individual stressors alone.</p></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>&#9632; ASSOCIATED CONTENT</head></div>
<div xmlns="http://www.tei-c.org/ns/1.0"><head>* s&#305; Supporting Information</head><p>The Supporting Information is available free of charge at h t t p s : / / p u b s . a c s . o r g / d o i / 1 0 . 1 0 2 1 / a c s e a r t h s p a c echem.2c00380.</p><p>Carbon and nitrogen content, solid-state <ref type="bibr">13</ref> C NMR spectra, all measurements conducted using solvent extraction, base hydrolysis, copper oxidation, PLFA analysis, and solid-state <ref type="bibr">13</ref>  </p></div><note xmlns="http://www.tei-c.org/ns/1.0" place="foot" xml:id="foot_0"><p>Downloaded via UNIV OF NEW HAMPSHIRE on March 30, 2023 at 20:47:50 (UTC).See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.</p></note>
			<note xmlns="http://www.tei-c.org/ns/1.0" place="foot" xml:id="foot_1"><p>https://doi.org/10.1021/acsearthspacechem.2c00380 ACS Earth Space Chem. 2023, 7, 609-622</p></note>
		</body>
		</text>
</TEI>
