This is the accepted version of the article: Tie, Liehua; Zhang, Shibin; Peñuelas, Josep; [et al.]. «Responses of soil C, N, and P stoichiometric ratios to N and S additions in a subtropical evergreen broad-leaved forest». Geoderma, Vol. 379 (Dec. 2020), art. 114633. DOI 10.1016/j.geoderma.2020.114633 This version is avaible at https://ddd.uab.cat/record/232210 under the terms of the license 1 Responses of soil C, N, and P stoichiometric ratios to N and S 2 additions in a subtropical evergreen broad-leaved forest 3 Liehua Tiea, Shibin Zhanga,b, Josep Peñuelasc,d, Jordi Sardansc,d, Shixing Zhoua, Junxi Hua, Congde 4 Huanga,* a 5 College of Forestry, Sichuan Agricultural University, Sichuan, 611130 Chengdu, China b 6 Sichuan Forestry and Grassland Inventory and Planning Institute, Sichuan, 611130 Chengdu, China c 7 CSIC, Unitat d’Ecologia Global CREAF-CEAB-CSIC-UAB, Edifici C, Universitat Autònoma de 8 Barcelona, Cerdanyola del Vallès, Catalonia, 08193 Barcelona, Spain d 9 CREAF, Cerdanyola del Vallès, Catalonia, 08193 Barcelona, Spain 10 11 Email addresses: tiefromchina@163.com (Liehua Tie), 275785445@qq.com (Shibin Zhang), 12 josep.penuelas@uab.cat (Josep Peñuelas), j.sardans@creaf.uab.cat (Jordi Sardans), djyzsx@163.com 13 (Shixing Zhou), 18681628120@163.com (Junxi Hu), lyyxq100@aliyun.com (Congde Huang). 14 15 * Corresponding author. Prof. Congde Huang, Email addresses: lyyxq100@aliyun.com 16 No. 211, Huimin Road, Wenjiang, Chengdu, Sichuan, 611130 China 17 1 18 Abstract 19 Acid deposition from the emission of nitrogen (N) and sulfur (S) has become an important factor 20 affecting the soil nutrient balance and biogeochemical cycling in terrestrial ecosystems. The average 21 levels of N and S deposition in the rainy area of southwestern China from 2008 to 2010 were 9.5 g N 22 m-2 y-1 and 19.3 g S m-2 y-1, respectively. External additions of N and S fertilizers combined with high 23 levels of acid deposition may affect the soil ecological stoichiometry in the region’s widely distributed 24 subtropical evergreen broad-leaved forest. Therefore, we investigated the responses of the soil 25 stoichiometric ratios and enzyme activities to added N (+N), added S (+S), added N and S (+NS), and 26 a control (Ctr) in the 0-20 cm layer in an evergreen broad-leaved forest in the rainy area of 27 southwestern China from April 2013 to April 2015. The results showed that the soil total N (TN) 28 concentration and N/P ratio were higher and the soil organic C (SOC) concentration and C/N ratio 29 were lower in the fertilization treatments than the Ctr, although N and S additions did not significantly 30 alter the soil total P (TP) concentration. The +N, +S, and +NS treatments increased the soil acid 31 phosphatase activity and reduced the soil invertase, cellulase, catalase, and polyphenol oxidase 32 activities. The +N and +NS treatments increased the soil urease activity and reduced soil peroxidase 33 activity. The +S treatment reduced the soil urease activity and did not alter soil peroxidase activity. N 34 and S additions had synergistic decreasing effects on the SOC concentration, C/N ratio, and soil 35 cellulose and catalase activities. Moreover, structural equation models identified that N and S additions 36 regulated the SOC, TN, and TP concentrations via shifting the activities of soil enzymes and the 37 pathways differed between N addition and S addition. In conclusion, N and S additions decreased the 38 SOC concentration, C/N ratio, and most soil C-cycle enzyme activities and increased the TN 39 concentration, N/P ratio, and soil acid phosphatase activity. All these results indicated that external N 2 40 and S additions combined with acid deposition increased soil N concentrations and exacerbated soil 41 C and P limitations in this forest. 42 Keywords: Acid deposition; Soil organic carbon; Nutrient limitation; Nitrogen-phosphorus imbalance; 43 Subtropical forest; Soil enzyme activities. 44 1. Introduction 45 Atmospheric acid deposition is an important component of global climate change (Bouwman et 46 al., 2002; Oulehle et al., 2013), and nitrogen (N) and sulfur (S) are the main elemental components of 47 atmospheric acid deposition (Anatolaki and Tsitouridou, 2007). The global atmospheric N and S 48 deposition in 2010 was estimated at approximately 123 Tg N y-1 and 84 Tg S y-1, respectively 49 (Galloway et al., 2008; Tan et al., 2018). Although the regional scales of atmospheric acid deposition 50 exhibit varying trends, acid deposition is still increasing in China (Gao et al., 2018; Tan et al., 2018). 51 As one of the regions with the highest level of acid deposition in the world, China accounts for 52 approximately 20% of the global atmospheric N deposition and approximately 17% of global 53 atmospheric S deposition (Du et al., 2015; Tan et al., 2018). Especially in some subtropical forests in 54 southern China, the acid deposition level is two to three times that of the national average (Du et al., 55 2015). Evidence suggests that inputs of N and S have affected soil nutrient imbalance (Gao et al., 2018; 56 Mori et al., 2019) and altered the biogeochemical cycles of terrestrial ecosystems (Güsewell and 57 Gessner, 2009; Lal, 2004; Sardans and Peñuelas, 2015). Therefore, external N and S fertilizers may 58 interact with high levels of atmospheric acid deposition to seriously threaten the soil nutrient balance 59 and alter biogeochemical cycles in these subtropical forests. 60 Soil stoichiometric ratios have been widely used to study the responses of soil ecosystem nutrient 61 balances and soil carbon (C), N, and phosphorus (P) biogeochemical cycles to the inputs of external 3 62 nutrient elements (Sardans and Peñuelas, 2012; van Dijk et al., 2012). External N addition generally 63 increases the availability of soil N (Magill et al., 2004; Michopoulos et al., 2004). However, the effects 64 of N addition on soil C and P concentrations and C/N, C/P, and N/P ratios vary with the level of 65 atmospheric deposition (Cleveland and Townsend, 2006), duration of fertilizer application (de Vries 66 et al., 2014), and type of ecosystem (Keeler et al., 2009). Thus, different results have been reported 67 for the effects of N addition on soil C and P concentrations, including increases (Cleveland and 68 Townsend, 2006; Marklein and Houlton, 2012), decreases (de Vries et al., 2014; Huang et al., 2016), 69 and no effects (Chen et al., 2018a; Hu et al., 2013). The effects of S addition on soil nutrient 70 concentrations and stoichiometric ratios are more uncertain than those of N addition, although S 71 addition can also affect nutrient dynamics (Fenn et al., 1999; Gao et al., 2018; Matejko et al., 2009). 72 Moreover, the inputs of external N can interact with S to alter the nutrient cycles of terrestrial 73 ecosystems (Keeler et al., 2009; Saiya-Cork et al., 2002), thereby increasing the uncertainty of the 74 responses of soil nutrient concentrations and stoichiometric ratios to N and S additions. 75 Soil enzymes are key in the control of biogeochemical cycles and soil nutrient concentrations 76 (Aragón et al., 2014). The activities of soil enzymes are very sensitive to the inputs of external nutrients 77 (Allison and Vitousek, 2005; Hu et al., 2013). Thus, the addition of N and S may alter the activities of 78 soil enzymes to regulate soil nutrient concentrations and stoichiometric ratios. For example, a meta- 79 analysis reported that N addition increased soil C concentrations in terrestrial ecosystems via shifting 80 the activities of soil enzymes (Chen et al., 2018b). However, the responses of soil nutrient 81 concentrations and stoichiometric ratios to N and S additions and how these responses are regulated 82 by soil enzymes in subtropical forests under high levels of acid deposition are poorly understood. 83 Therefore, investigating how N and S additions alter soil nutrient concentrations and stoichiometric 4 84 ratios via shifting soil enzyme activities will improve our understanding of the mechanism underlying 85 the effect of adding N and S on the concentrations and ratios. 86 The rainy area of southwestern China has received considerable acid deposition and represents 87 one of the areas of China with the greatest levels of deposition (Tu et al., 2013). Subtropical evergreen 88 broad-leaved forests are the most widely distributed and representative type of forest in this region. 89 They provide excellent experimental sites for examining the responses of soil stoichiometric ratios 90 and soil enzyme activities to N and S additions in subtropical forests under high levels of acid 91 deposition. In this study, we conducted a field experiment in a subtropical evergreen broad-leaved 92 forest in the rainy area of southwestern China for two years. We hypothesized that (i) N and S additions 93 would affect the concentrations and stoichiometric ratios of soil nutrients via shifting the activities of 94 soil enzymes in this forest. N and S additions can have different effects on the activities of soil enzymes 95 (Hu et al., 2013; Lv et al., 2014), and the mechanisms underlying the effects of N and S additions on 96 soil nutrient concentrations and stoichiometric ratios may be different. Therefore, we also 97 hypothesized that (ii) the responses of soil nutrient concentrations and stoichiometric ratios to N 98 addition and S addition would differ. In addition, due to the strong coupling between N and S cycles 99 (Gao et al., 2018), we hypothesized that (iii) N and S additions would have interactive effects on the 100 nutrient concentrations and stoichiometric ratios in the study subtropical forest. 101 2. Materials and methods 102 2.1. Study site 103 The rainy area of southwestern China is an ecotone between the Sichuan Plateau and the Sichuan 104 Basin that spans 70 km from east to west and 450 km from north to south, and the total area is 105 approximately 25,000 km2 (Zhou et al., 2018). The Sichuan Plateau and Basin differ greatly in 5 106 elevation. Warm and humid air from the Sichuan Basin and the city of Chongqing (east of the basin) 107 usually condenses into rain in this rainy area as the elevation increases, thereby providing abundant 108 rainfall (Tu et al., 2013; Xu et al., 2013). Rapid agricultural intensification, industrial and urban 109 development, and increased fuel use for transportation and energy production in the Sichuan Basin 110 and the city of Chongqing have resulted in dramatic increases in N and S pollution since the 1980s 111 (Xu et al., 2013). N and S pollutants are transported to this rainy area and deposited on the ground 112 with rain due to the special climate and topography of the area (Tu et al., 2013; Xu et al., 2013). From 113 2008 to 2010, the annual average wet N and S depositions were 9.5 g N m-2 y-1 and 19.3 g S m-2 y-1, 114 respectively (Lin and Hu, 2011; Tu et al., 2013). The pH of the precipitation ranged from 3.9 to 5.0, 115 and the annual average was 4.4 (Lin and Hu, 2011). 116 Ya’an, Sichuan Province is located in the center of this rainy area. Thus, we selected a subtropical 117 evergreen broad-leaved forest on Jinfengsi Mountain (30°02′N, 103°03′E; 800 m a.s.l.) in Ya’an in 118 October 2012 to study the effects of N and S additions on the concentrations and stoichiometric ratios 119 of soil nutrients. The study site has a subtropical humid monsoon climate. The annual average 120 temperature is 16.1 °C, and the annual average rainfall is 1772 mm. Approximately 70% of the rain 121 falls from May to October, and < 20 mm falls in December and January. Both the temperature and 122 rainfall are lowest in January and highest in August. The stand age at the study site is approximately 123 40 years. The dominant tree species are Lindera megaphylla, Quercus serrata, Choerospondias 124 axillaris, and Eurya japonica. The main species in the undergrowth is Dicranopteris linearis. The soil 125 is classified as a Lithic Dystrudept (according to USDA Soil Taxonomy) derived from purple 126 sandstone and shale. The average thickness of the surface organic layer is approximately 2 cm. The 127 average soil depth to bedrock is approximately 80 cm. The surface soil horizon (0-20 cm) organic C 6 128 was 16.2±0.7 (mean ± standard deviation) g kg-1, total N (TN) was 1.60±0.06 g kg-1, and pH was 129 4.50±0.09 (H2O extraction) before the experimental treatments began. The annual temperature and 130 water content of the surface soil horizon (0-20 cm) were 11.9±0.4 °C and 27.9±1.1% from April 2013 131 to April 2015, respectively (Fig. S1). 132 2.2. Experimental design 133 Considering that the levels of atmospheric N and S depositions in this rainy area are very high 134 (Tu et al., 2013; Xu et al., 2013) and expected to continue increasing until at least 2030 (Du et al., 135 2015) and drawing on the experience of N addition in this region (Peng et al., 2020; Tu et al., 2013; 136 Zhou et al., 2018), 15 g N m-2 y-1 and 30 g S m-2 y-1 additions were tested in this study. The treatments 137 consisted of a control (Ctr, 0 g N m-2 y-1 and 0 g S m-2 y-1), added N (+N, 15 g N m-2 y-1), added S (+S, 138 30 g S m-2 y-1), and added N and S (+NS, 15 g N m-2 y-1 and 30 g S m-2 y-1). Urea and sodium sulfate 139 were used as the N and S fertilizers, respectively. 140 Twelve plots (5 × 5 m) were established in the study forest in March 2013. The distance between 141 each plot was > 5 m. To reduce errors caused by slope, topographic position, and nutrient heterogeneity, 142 the twelve plots were divided into three blocks perpendicular to the maximum slope. The four 143 treatments were randomly distributed to plots inside each block, and each treatment was repeated three 144 times. The N and S fertilizers were divided into 48 equal parts and applied every half month from 145 April 2013 to April 2015. The fertilizers were dissolved in 2 L of water for each application and then 146 sprayed evenly on the soil surface. The Ctr plots were sprayed with 2 L of water without fertilizer to 147 avoid experimental error. Based on the rainfall and N and S concentrations in the precipitation, we 148 estimated that approximately 7.75 g N m-2 y-1 and 15.8 g S m-2 y-1 were deposited onto the soil surface 149 from natural rain during the study period. 7 150 2.3. Soil sampling and analysis 151 Soil samples were collected eight times: July and October 2013; January, April, July, and October 152 2014; and January and April 2015. Moss and leaf litter were removed from the soil surface, and five 153 soil samples (0-20 cm) were collected in each plot using a soil auger (diameter of 5 cm). The five 154 samples from each plot were then fully mixed and transported to the laboratory. In the laboratory, the 155 roots and gravel in the samples were removed by tweezers. The soil sample was homogenized and 156 then divided into two subsamples. One subsample was sieved (< 2 mm) and stored at 4 °C for 157 measuring the activities of soil enzymes within one week. The other subsample was air-dried and 158 sieved (< 2 mm) for measuring the concentrations of soil organic C (SOC), TN, total P (TP), and pH. 159 The SOC concentration was determined using dichromate oxidation-ferrous sulfate titration 160 (Schinner et al., 1996), the TN concentration was determined by Kjeldahl digestion (Allen et al., 1974), 161 the TP concentration was determined by molybdenum-antimony colorimetry (Allen et al., 1974), and 162 the soil pH was determined by a glass electrode in aqueous extracts (Allen et al., 1974). Soil enzyme 163 activities were determined by spectrophotometrically. The activity of invertase (E.C. 3.2.1.26) was 164 determined using sucrose as the substrate (Ohshima et al., 2007), the activity of urease (E.C. 3.5.1.5) 165 was determined using urea as the substrate (Kandeler and Gerber, 1988), the activity of acid 166 phosphatase (E.C. 3.1.3.2) was determined using para-nitrophenyl phosphate as the orthophosphate 167 monoester analog substrate (Schinner et al., 1996; Tabatabai, 1982), the activity of cellulase (E.C. 168 3.2.1.4) was determined using carboxymethylcellulose sodium as the substrate (Ghose, 1987), the 169 activity of catalase (E.C. 1.11.1.6) was determined using hydrogen peroxide as the substrate (Tabatabai, 170 1982), and the activities of peroxidase (E.C. 1.11.1.7) and polyphenol oxidase (E.C. 1.10.3.2) were 171 determined using L-3,4-dihydroxyphenylalanine as the substrate (Keeler et al., 2009). The detailed 8 172 measurements of soil enzyme activities are shown in Table S1. All measurements were repeated three 173 times. 174 2.4. Data analysis 175 The soil stoichiometric ratios in this text are mass-based, e.g., C/N is the ratio of SOC to TN 176 concentrations (g kg-1), C/P is the ratio of SOC to TP concentrations, and N/P is the ratio of TN to TP 177 concentrations. 178 The variables were statistically analyzed using SPSS 20.0 for Windows (SPSS Inc., Chicago, 179 USA). All variables were first tested for a normal distribution and homogeneity of variance using the 180 one-sample Kolmogorov-Smirnov test and Levene’s test, respectively. Variables with non-normal or 181 unequal variances, i.e., the C/N ratio and soil cellulose, peroxidase, and polyphenol oxidase activities, 182 were transformed using the Box-Cox method. A two-way ANOVA was performed to examine the 183 effects of N and S additions on the SOC, TN, and TP concentrations; soil C/N, C/P, and N/P ratios; 184 and soil enzyme activities at each sampling time. To understand how N and S additions affected the 185 soil nutrient concentrations, stoichiometric ratios, pH value, and enzyme activities during the whole 186 fertilization process, linear mixed effects models employing the restricted maximum likelihood 187 estimation method were used to examine the effects of N addition, S addition, and their interactions 188 on the mean SOC, TN, and TP concentrations; soil C/N, C/P, and N/P ratios; soil pH; and soil enzyme 189 activities during two years of fertilization. We included N addition, S addition, and sampling time as 190 fixed factors and blocks and plots nested in blocks as random factors. The Bonferroni adjustment 191 method was applied to the confidence intervals and significance values to account for multiple 192 comparisons. The pairwise relationships among the SOC, TN, and TP concentrations and the soil 193 stoichiometric ratios in the different treatments and control were examined by using distinct regression 9 194 approaches, such as curve estimation with the unary linear regression model, logarithmic regression 195 model, quadratic regression model, and exponential regression model. Then, the optimal model was 196 chosen in each case. To understand how N and S additions directly and indirectly affect soil nutrient 197 concentrations, we also used the structural equation model (SEM) in AMOS 22.0 (IBM SPSS, Chicago, 198 USA) to examine the effects of the N and S additions on the soil nutrient concentrations and to compare 199 the relative contributions of the soil enzyme activities to the SOC, TN, and TP concentrations. 200 Coefficients were estimated using the maximum likelihood method, and confidence intervals were set 201 at 95%. Differences were deemed significant at P < 0.05 for all analyses. 202 To understand the overall differences in the soil nutrient concentrations, stoichiometric ratios, 203 and enzyme activities among different treatments, we performed multivariate statistical analyses using 204 a general discriminant analysis (GDA) to determine the overall differences in SOC, TN and TP 205 concentrations; C/N, C/P, and N/P ratios; and soil enzyme activities among the Ctr and different 206 fertilization treatments. We also assessed the component of the variance caused by the sampling time 207 as an independent categorical variable. This analysis used the squared Mahalanobis distance statistic 208 that depends on the Euclidean distance in the model between two sets of samples: as the sets became 209 closer and less different and the squared Mahalanobis distance decreased; and as the sets became more 210 distant and different and the squared Mahalanobis distance increased (De Maesschalck et al., 2000). 211 GDA is thus an appropriate tool for identifying the variables most responsible for the differences 212 among groups while controlling for the component of the variance caused by other categorical 213 variables, e.g., sampling time. The GDA was performed using Statistica 8.0 (StatSoft, Inc., Tulsa, 214 USA). 215 3. Results 10 216 3.1. Effects of N and S additions on soil nutrient concentrations and soil pH 217 The SOC concentration was lower in the +N, +S, and +NS than the Ctr treatment after fifteen 218 months (P < 0.05; Fig. S2a), and the TN concentration was higher in the fertilization treatments than 219 the Ctr after eighteen months (P < 0.05; Fig. S2b). In contrast, the TP concentration in all treatments 220 ranged from 0.4 to 0.6 g kg-1 (Fig. S2c). The +N, +S, and +NS decreased the mean SOC concentration 221 by 14.0%, 10.5%, and 19.7%, respectively (P < 0.05; Fig. 1a), increased the mean TN concentration 222 by 15.4%, 15.3%, and 34.3%, respectively (P < 0.05; Fig. 1b), but did not affect the mean TP 223 concentration (P > 0.05; Fig. 1c). The mean TN concentration was higher in the +NS treatment than 224 the +N and +S treatments (P < 0.05). N and S additions interactively decreased the SOC concentrations 225 (P < 0.05; Table 1). The soil pH in the +N, +S, and +NS were lower than that in the Ctr by 3.07%, 226 1.98%, and 4.58%, respectively (P < 0.05; Fig. S3). 227 3.2. Effects of N and S additions on soil stoichiometric ratios 228 The C/N ratio was lower in the +N, +S, and +NS than the Ctr after fifteen months (P < 0.05; Fig. 229 S4a), and the N/P ratio was higher in the fertilization treatments than the Ctr after twelve months (P < 230 0.05; Fig. S4c). The +N, +S, and +NS reduced the mean C/N ratio by 27.1%, 22.7%, and 39.8%, 231 respectively (P < 0.05; Fig. 1d), and increased the mean N/P ratio by 17.7%, 22.4%, and 41.2%, 232 respectively (P < 0.05; Fig. 1f). The C/N ratio was lower and the N/P ratio was higher in the +NS than 233 the +N and +S (P < 0.05). Moreover, the mean C/P ratio was reduced in the +N and +NS (P < 0.01; 234 Fig. 1e) but was not altered by the +S (P > 0.05). The N and S additions interactively decreased the 235 soil C/N ratios (P < 0.01; Table 1). 236 3.3. Effects of N and S additions on soil enzyme activities 237 The activities of the soil enzymes were seasonally dynamic, and the responses of the activities 11 238 varied among treatments (Fig. S5a-g). The +N, +S, and +NS treatments reduced the mean activities of 239 soil invertase, cellulose, catalase, and polyphenol oxidase (P < 0.05; Fig. 2a, d, e, and g) and increased 240 the mean activity of acid phosphatase (P < 0.05; Fig. 2c). The +N and +NS increased the mean activity 241 of soil urease by 10.6% and 8.9%, respectively (P < 0.05; Fig. 2b), and reduced the mean activity of 242 soil peroxidase by 12.3% and 14.8%, respectively (P < 0.05; Fig. 2f). The +S decreased the mean 243 activity of soil urease by 9.8% (P < 0.05; Fig. 2b) but did not alter the mean activity of soil peroxidase 244 (P > 0.05; Fig. 2f). N and S additions interactively decreased the activities of cellulase and catalase (P 245 < 0.05; Table 2). 246 3.4. Relationships among the SOC, TN, and TP concentrations and soil stoichiometric ratios 247 The SOC and TN concentrations, SOC concentration and N/P ratio were negatively and linearly 248 correlated in the samples from the +N, +S, and +NS treatments (P < 0.05; Fig. 3a and b) but 249 quadratically correlated in the Ctr (P < 0.05). The TN and C/P were negatively and linearly correlated 250 in the +N, +S, and +NS (P < 0.05; Fig. 3d) but were not significantly correlated in the Ctr (P > 0.05). 251 The SOC and TP concentrations were not significantly correlated in the Ctr, +N, or +S (P > 0.05; Fig. 252 3c) but were positively correlated in the +NS (P < 0.05). 253 3.5. Contributions of the enzyme activities to the SOC, TN, and TP concentrations 254 Although N and S additions directly increased the SOC and TN concentrations, their indirect 255 effects on nutrient concentrations through soil enzyme activities were the dominant effects (Table 3). 256 N and S additions both decreased SOC concentration, with N addition decreasing the activities of 257 invertase, cellulase, and peroxidase and increasing the activities of urease and acid phosphatase (Fig. 258 4a) and S addition decreasing the activities of invertase, urease, and peroxidase and increasing the acid 259 phosphatase activity (Fig. 4b). N and S additions mainly increased the acid phosphatase activity and 12 260 decreased the peroxidase activity to increase the TN concentration (Fig. 4c and d). N addition also 261 increased the soil TN concentration by improving the soil urease activity, but this result cannot be 262 taken into account because the added N was in the form of urea. N and S additions did not directly 263 affect the soil TP concentration but indirectly affected the TP concentration by altering the soil enzyme 264 activities (Fig. 4e and f, Table 3). The standardized total effects of N and S additions on the soil TP 265 concentrations were 0.0022 and -0.157, respectively (P < 0.05; Table 3). 266 3.6. General discriminant analysis 267 The 2D layout represented by the two first roots (explaining 80.6% of the total variance) was 268 clearly consistent with the observations from the univariate and SEM analyses (Fig. 5). The main 269 variables loading on Root 1 were soil TN concentration, N/P ratio, and acid phosphatase activity, which 270 were higher towards the S addition treatments, and soil invertase activity and C/N ratio, which were 271 lower towards the S addition treatments. The main variables loading on Root 2 were soil TN 272 concentration, urease activity, and N/P ratio, which were higher towards the N addition treatments, 273 and SOC concentration, C/N and C/P ratios, and peroxidase and cellulase activities, which were lower 274 towards the N addition treatments. N and S additions had synergistic increasing effects on the soil TN 275 concentration, N/P ratio, and acid phosphatase activity but synergistic decreasing effects on the SOC 276 concentration, C/N and C/P ratios, and invertase, catalase, and peroxidase activities. The main effects 277 of the variables in the GDA showed that soil TN concentration, soil C/N ratio, and all of the soil 278 enzyme activities were the main variables responsible for the overall differences among soils receiving 279 distinct treatments (Table S2 and S3). This information reinforces the results observed in the univariant 280 analyses. 281 4. Discussion 13 282 4.1. Effects of N and S additions on SOC concentration 283 The responses of SOC concentration to N addition vary with the fertilization level and duration 284 of application (Cleveland and Townsend, 2006; Lu et al., 2013). Short-term (< 3 years) high levels of 285 N input (> 10 g N m-2 y-1) generally decrease SOC concentration (Cleveland and Townsend, 2006; de 286 Vries et al., 2014). Similarly, N addition in our study (15 g N m-2 y-1) also decreased the SOC 287 concentration in the 0-20 cm soil layer during the two years of fertilization (Fig. 1a). Two potential 288 mechanisms may account for the effect of N addition on SOC concentration. First, the associated 289 decrease in SOC concentration may be related to the reduction of most soil C-cycle enzyme activities 290 (Fig. 4a), e.g., soil invertase, cellulase, and peroxidase activities, which strongly suggests a decrease 291 in soil biological activity (Allison and Vitousek, 2005; Aragón et al., 2014). This decrease in biological 292 activity may reduce the potential plant litter decomposition rate and transformation rate into SOC, 293 which results in lower soil C inputs. A meta-analysis also showed that N addition reduced most soil 294 C-cycle enzyme activities when the level of fertilization reached 15 g N m-2 y-1 (Jian et al., 2016), 295 which supported the results in our study. Second, the relatively high levels of both N addition and 296 atmospheric N deposition can exacerbate soil C leaching, which may account for the decrease of SOC 297 concentration. For example, Scott et al. (2015) reported that N addition (10 g N m-2 y-1) increased the 298 amount of soil dissolved organic C leaching by 70%. 299 The S addition in our study reduced SOC concentration of the 0-20 cm soil layer (Fig. 1a). In 300 general, lower concentrations of soil fulvic acid, total carbohydrates, and total organic C were found 301 in the soil samples with lower pH (Hay et al., 1985). We found that the addition of S lowered the soil 302 pH (Fig. S3), which may have caused a decrease in SOC concentration. Moreover, S interacted with 303 N to reduce SOC concentration (Table 1). On the one hand, the decrease of soil pH value was stronger 14 304 under the +NS treatment than the +N or +S treatments (Fig. S3). Thus, the +NS treatment may have a 305 stronger inhibitory effect on the biogeochemical cycle of soil C than +N or +S treatments (Keeler et 306 al., 2009; Zhang et al., 2018). On the other hand, the N and S additions in our study synergistically 307 decreased the activities of cellulase and catalase (Table 2), which may have synergistically reduced 308 the return of plant organic C back to soil (Allison and Vitousek, 2005), resulting in a synergistic 309 decreasing effect on SOC concentration (Fig. 5). 310 Our previous study (Tie et al., 2018) reported that N and S additions inhibited the decomposition 311 of leaf litter in the same forest and that organic matter accumulated in the layers of leaf litter. The 312 organic C in the leaf litter could not return to the soil with short-term fertilization, which supported 313 our inferences above. Therefore, short-term N and S additions may have interacted the high levels of 314 atmospheric acid deposition to inhibit the C cycle, stimulate C sequestration in the litter layer, and 315 decrease the C concentration in the 0-20 cm soil layer of this study forest. These findings are in line 316 with those of other studies (Magill and Aber, 1998; Xu et al., 2017). However, Hu et al. (2013) reported 317 that N (3.0 g N m-2 y-1) and S (3.0 g S m-2 y-1) additions did not change the SOC concentration in a 318 boreal forest, where the levels of atmospheric N and S deposition were 0.08 and 0.11 g m-2 y-1, 319 respectively. The effects of external N and S inputs on soil C concentration vary with their levels 320 (Cleveland and Townsend, 2006), and the levels of atmospheric deposition and fertilization differed 321 greatly between our site and the site studied by Hu et al. (2013), which may account for the different 322 results between the two studies. In addition, the decrease in SOC with time under N and S additions 323 (Fig. 1a) supports the idea that their additions may have cumulative effects on SOC concentration (de 324 Vries et al., 2014; Mori et al., 2019); thus, medium- and long-term N and S additions are likely to 325 accelerate the reduction of SOC concentration in our study forest. 15 326 4.2. Effects of N and S additions on soil TN concentration 327 The addition of N generally directly increases external N and thus increases the soil TN 328 concentration (Magill et al., 2004; Michopoulos et al., 2004). The direct effects of N addition on soil 329 TN concentration in our study were consistent with previous studies (Magill et al., 2004; Michopoulos 330 et al., 2004). Moreover, the addition of N can indirectly alter the concentration of soil N via shifting 331 soil enzyme activities (Allison, 2005; Keeler et al., 2009). SEM analyses showed that N addition led 332 to increased soil TN concentration, at least partially, by indirectly increasing the soil acid phosphatase 333 activity and decreasing the soil peroxidase activity (Fig. 4c). Peroxidases can oxidize phenols, amines, 334 and hydrocarbons (Keeler et al., 2009). N addition in our study decreased peroxidase activity (Fig. 2f), 335 which could inhibit the decomposition of nitrogenous phenols, amines, and hydrocarbons and may 336 lead to the accumulation of N. Organic P and organic N are mainly present in carbonaceous organic 337 matter (Prescott, 2010). Soil acid phosphatase can hydrolyze P groups from various organic-P 338 substrates in organic matter (Saiya-Cork et al., 2002) and may indirectly promote the hydrolysis of 339 organic N (Jian et al., 2016). The increase of soil acid phosphatase activity under N addition in our 340 study therefore may have promoted the hydrolysis of organic N. Moreover, N addition increased the 341 negative correlations between SOC and TN, SOC and N/P, and TN and C/P (Fig. 3a, b, and d), which 342 may also be related to the increased soil acid phosphatase activity and decreased soil peroxidase 343 activity. 344 The addition of sodium sulfate can increase the leaching of soil exchangeable cations (e.g., K+, 345 Mg2+, and Ca2+) because Na+ can displace the rest of soil cations from the exchange complex, which 346 usually alters microbial and enzyme activities to influence N cycling in forest ecosystems (Gao et al., 347 2018; Mori et al., 2019). In our study, the increased TN concentration was stronger under the +NS 16 348 treatment than the +N or +S treatments (Fig. 1b). S addition thus assisted the N fertilizer in increasing 349 the soil TN concentrations at the 0-20 cm soil layer, which may be related to S addition reducing the 350 soil N output (e.g., N2O emission). For example, Gao et al. (2014) reported that Na2SO4 addition 351 decreased N2O emissions by 79% from the soil in the Tibetan Plateau, China. Moreover, ammonium 352 sulfate could be formed in the soil when Na2SO4 was added in combination with urea, which is 353 important because ammonium sulfate releases ammonium more slowly than urea and leads to a greater 354 reduction in the potential volatilization and leaching of ammonium compared with urea (Chien et al., 355 2011). The effects of N and S additions on TN concentration in this forest may thus be more 356 complicated than previously thought. 357 Most of the external N inputs generally leach and erode into rivers. For example, Fang et al. 358 (2009) reported that 25-66% of the input of external N leached and eroded into rivers in three seasonal 359 monsoon subtropical forests (evergreen broad-leaved forest, needle-leaved forest, and needle- and 360 broad-leaved mixed forest) with the same annual average rainfall of 1927 mm (approximately 75% of 361 rain falls from March to August). Our study site was also in a seasonal monsoon subtropical area, with 362 an annual average rainfall of 1772 mm (approximately 70% of rain falls from May to October). 363 Although S addition assisted the N fertilizer in increasing the soil TN concentration, some external N 364 may still have leached and eroded. Previous studies found that the Yangtze River has undergone 365 eutrophication mainly due to increases in the levels of nitrogenous compounds in the water (Chi et al., 366 2017). Our study site is in the upper reaches of the Yangtze River; therefore, the eutrophication of the 367 Yangtze River may become more serious with high levels of urea and sodium sulfate addition. More 368 studies, however, are needed. 369 4.3. Effects of N and S additions on soil TP concentration 17 370 P is an essential element in all life and responsible for a considerable proportion of the growth 371 and development of vegetation (Paytan and McLaughlin, 2007; Turner, 2008). However, the effects of 372 external N and S inputs on the soil P cycle are still unclear. The N and S additions in our study did not 373 significantly alter the soil TP concentration (Fig. 1c) but rather increased the soil acid phosphatase 374 activity at the 0-20 cm soil layer (Fig. 2c), strongly suggesting that the demand for P in the 0-20 cm 375 soil ecosystem was increased (Peñuelas et al., 2013). Our previous study (Tie et al., 2018) also reported 376 that N and S additions slowed the release of P during litter decomposition, which may have reduced P 377 return to soil and strongly suggests a soil P limitation at the 0-20 cm soil layer. 378 The soil P cycle is a slow and complex biogeochemical process (Paytan and McLaughlin, 2007; 379 Peñuelas et al., 2013), so the TP concentration at 0-20 cm did not respond strongly to the short-term 380 additions of N and S. However, after 21 months of fertilization, N and S additions slightly decreased 381 the TP concentration (Fig. S2c), suggesting that long-term N and S additions potentially increase the 382 P demand and medium- and long-term additions may decrease the TP concentration in this forest. 383 4.4. Effects of N and S additions on the soil C/N, C/P, and N/P ratios 384 Soil stoichiometric ratios are at the core of C, N, and P biogeochemistry (Sardans and Peñuelas, 385 2015), which regulates the above- and belowground nutrient cycles of forest ecosystems (Güsewell 386 and Gessner, 2009; Peng et al., 2019). Soil stoichiometric ratios are inherently stable until external 387 nutrients are added (Huang et al., 2018). In our study, N and S additions decreased the soil C/N ratio 388 and increased the soil N/P ratio (Fig. 1d and f), which is consistent with the findings reported in other 389 studies (Huang et al., 2018; van Dijk et al., 2012). Changes in soil stoichiometric ratios can alter 390 nutrient cycles in plant-soil systems (Peng et al., 2019; Sardans and Peñuelas, 2015), which may affect 391 tree health and forest services (Aherne and Posch, 2013; Yuan and Chen, 2015). 18 392 Soil C and P limitations are generally caused by decreases in the soil C/N ratio and increases in 393 the soil N/P ratio, respectively (Peng et al., 2019; Zechmeister-Boltenstern et al., 2015). The lower soil 394 C/N ratio and higher soil N/P ratio in the soil samples from the N and S addition treatments in our 395 study indicated that the N and S additions may have aggravated the soil C and P limitations at the 0- 396 20 cm soil layer of the forest (Cleveland and Townsend, 2006). Moreover, microorganisms and plants 397 generally secrete more phosphatase in the case of P limitation (Zechmeister-Boltenstern et al., 2015). 398 The N and S additions in our study synergistically increased the acid phosphatase activity (Fig. 2c), 399 which reinforced the overall results strongly suggesting a clear exacerbation of P as the main limiting 400 factor. Atmospheric acid deposition in the rainy area of southwestern China is projected to remain high 401 until at least 2030 (Du et al., 2015). Therefore, the addition of N and S fertilizers combined with high 402 levels of atmospheric acid deposition may continue increasing the soil N concentrations and 403 exacerbating the soil C and P limitations at the 0-20 cm soil layer of these subtropical forests in the 404 rainy area of southwestern China in the future. 405 5. Conclusions 406 N and S additions decreased the SOC concentration, soil C/N ratio, and most soil C-cycle enzyme 407 activities and increased the soil TN concentration, soil N/P ratio, and soil acid phosphatase activity at 408 the 0-20 cm soil layer of the forest. These results indicated that N and S additions combined with acid 409 deposition increased soil N concentrations and exacerbated soil C and P limitations in this forest. N 410 and S additions regulated SOC, TN, and TP concentrations via shifting the activities of soil enzymes. 411 The responses of the soil C, N, and P concentrations and stoichiometric ratios to the N and S additions 412 differed because the effects of their additions on the activities of soil urease, cellulase, catalase, and 413 peroxidase differed. Moreover, the N and S additions synergistically decreased the SOC concentration, 19 414 C/N ratio, and soil cellulose and catalase activities, which suggested that N and S additions may have 415 synergistically exacerbated soil C limitation at the 0-20 cm soil layer. The responses of soil nutrient 416 concentrations and stoichiometric ratios to N and S additions, however, are complex. Therefore, the 417 long-term effects of N and S additions on soil C, N, and P stoichiometric ratios in this forest should 418 be investigated. 20 419 Acknowledgments 420 This work is supported by the Crop Breeding Research Project of the ‘12th Five-year Plan’ of Sichuan 421 Province [2011NZ0098-10] and the Sci-tech Project of the ‘12th Five-year Plan’ of China 422 [2010BACO1A11]. Josep Peñuelas and Jordi Sardans are financially supported by the European 423 Research Council Synergy grant [ERC-SyG-2013-610028, IMBALANCE-P]. Liehua Tie appreciates 424 the financial support from the China Scholarship Council [No. 201906910006]. 425 426 Competing interests 427 The authors declare that they have no competing financial interests. 428 21 429 References 430 Aherne, J., Posch, M., 2013. Impacts of nitrogen and sulphur deposition on forest ecosystem services 431 in Canada. Curr. Opin. Environ. Sustain. 5, 108-115. 432 https://doi.org/10.1016/j.cosust.2013.02.005. 433 Allen, S.E., Grimshaw, H.M., Parkinson, J.A., Quarmby, C., 1974. Chemical analysis of ecological 434 materials. Blackwell Scientific Publications, Oxford, UK and London, UK. 435 Allison, S.D., Vitousek, P.M., 2005. Responses of extracellular enzymes to simple and complex 436 nutrient inputs. Soil Biol. Biochem. 37, 937-944. https://doi.org/10.1016/j.soilbio.2004.09.014. 437 Anatolaki, C., Tsitouridou, R., 2007. Atmospheric deposition of nitrogen, sulfur and chloride in 438 Thessaloniki, Greece. Atmos. Res. 85, 413-428. https://doi.org/10.1016/j.atmosres.2007.02.010. 439 Aragón, R., Sardans, J., Peñuelas, J., 2014. Soil enzymes associated with carbon and nitrogen cycling 440 in invaded and native secondary forests of northwestern Argentina. Plant Soil 384, 169-183. 441 https://doi.org/10.1007/s11104-014-2192-8. 442 Bouwman, A.F., Van Vuuren, D.P., Derwent, R.G., Posch, M., 2002. A global analysis of acidification 443 and eutrophication of terrestrial ecosystems. Water Air Soil Pollut. 141, 349-382. 444 https://doi.org/10.1023/a:1021398008726. 445 Chen, H., Chen, M.L., Li, D.J., Mao, Q.G., Zhang, W., Mo, J.M., 2018a. Responses of soil phosphorus 446 availability to nitrogen addition in a legume and a non-legume plantation. Geoderma 322, 12-18. 447 https://doi.org/10.1016/j.geoderma.2018.02.017. 448 Chen, J., Luo, Y.Q., van Groenigen, K.J., Hungate, B.A., Cao, J.J., Zhou, X.H., Wang, R.W., 2018b. 449 A keystone microbial enzyme for nitrogen control of soil carbon storage. Sci. Adv. 4, eaaq1689. 450 https://doi.org/10.1126/sciadv.aaq1689. 451 Chi, L.B., Song, X.X., Yuan, Y.Q., Wang, W.T., Zhou, P., Fan, X., Cao, X.H., Yu, Z.M., 2017. 452 Distribution and key influential factors of dissolved oxygen off the Changjiang River Estuary 453 (CRE) and its adjacent waters in China. Mar. Pollut. Bull. 125, 440-450. 454 https://doi.org/10.1016/j.marpolbul.2017.09.063. 455 Chien, S.H., Gearhart, M.M., Villagarcia, S., 2011. Comparison of ammonium sulfate with other 456 nitrogen and sulfur fertilizers in increasing crop production and minimizing environmental 457 impact: a review. Soil Sci. 176, 327-335. https://doi.org/10.1097/SS.0b013e31821f0816. 22 458 Cleveland, C.C., Townsend, A.R., 2006. Nutrient additions to a tropical rain forest drive substantial 459 soil carbon dioxide losses to the atmosphere. Proc. Natl. Acad. Sci. U. S. A. 103, 10316-10321. 460 https://doi.org/10.1073/pnas.0600989103. 461 De Maesschalck, R., Jouan-Rimbaud, D., Massart, D.L., 2000. The Mahalanobis distance. Chemom. 462 Intell. Lab. Syst. 50, 1-18. https://doi.org/10.1016/s0169-7439(99)00047-7. 463 de Vries, W., Du, E.Z., Butterbach-Bahl, K., 2014. Short and long-term impacts of nitrogen deposition 464 on carbon sequestration by forest ecosystems. Curr. Opin. Environ. Sustain. 9-10, 90-104. 465 https://doi.org/10.1016/j.cosust.2014.09.001. 466 Du, E., de Vries, W., Liu, X., Fang, J., Galloway, J.N., Jiang, Y., 2015. Spatial boundary of urban 'acid 467 islands' in southern China. Sci. Rep. 5, 12625. https://doi.org/10.1038/srep12625. 468 Fang, Y.T., Gundersen, P., Mo, J.M., Zhu, W.X., 2009. Nitrogen leaching in response to increased 469 nitrogen inputs in subtropical monsoon forests in southern China. For. Ecol. Manag. 257, 332- 470 342. https://doi.org/10.1016/j.foreco.2008.09.004. 471 Fenn, M.E., De Bauer, L.I., Quevedo-Nolasco, A., Rodriguez-Frausto, C., 1999. Nitrogen and sulfur 472 deposition and forest nutrient status in the Valley of Mexico. Water Air Soil Pollut. 113, 155-174. 473 https://doi.org/10.1023/a:1005033008277. 474 Galloway, J.N., Townsend, A.R., Jan Willem, E., Mateete, B., Zucong, C., Freney, J.R., Martinelli, 475 L.A., Seitzinger, S.P., Sutton, M.A., 2008. Transformation of the nitrogen cycle: recent trends, 476 questions, and potential solutions. Science 320, 889-892. 477 https://doi.org/10.1126/science.1136674. 478 Gao, Y., Chen, H., Zeng, X., 2014. Effects of nitrogen and sulfur deposition on CH4 and N2O fluxes 479 in high-altitude peatland soil under different water tables in the Tibetan Plateau. Soil Sci. Plant 480 Nutr. 60, 404-410. https://doi.org/10.1080/00380768.2014.893812. 481 Gao, Y., Ma, M., Yang, T., Chen, W., Yang, T., 2018. Global atmospheric sulfur deposition and 482 associated impaction on nitrogen cycling in ecosystems. J. Clean. Prod. 195, 1-9. 483 https://doi.org/10.1016/j.jclepro.2018.05.166. 484 Ghose, T.K., 1987. Measurement of cellulase activities. Pure Appl. Chem. 59, 257-268. 485 https://doi.org/10.1351/pac198759020257. 486 Güsewell, S., Gessner, M.O., 2009. N:P ratios influence litter decomposition and colonization by fungi 487 and bacteria in microcosms. Funct. Ecol. 23, 211-219. https://doi.org/10.1111/j.1365- 23 488 2435.2008.01478.x. 489 Hay, G.W., James, J.H., Vanloon, G.W., 1985. Solubilization effects of simulated acid rain on the 490 organic matter of forest soil; preliminary results. Soil Sci. 139, 422-430. 491 https://doi.org/10.1097/00010694-198505000-00007. 492 Hu, Y.L., Jung, K., Zeng, D.H., Chang, S.X., 2013. Nitrogen- and sulfur-deposition-altered soil 493 microbial community functions and enzyme activities in a boreal mixedwood forest in western 494 Canada. Can. J. For. Res. 43, 777-784. https://doi.org/10.1139/cjfr-2013-0049. 495 Huang, J.Y., Wang, P., Niu, Y.B., Yu, H.L., Ma, F., Xiao, G.J., Xu, X., 2018. Changes in C:N:P 496 stoichiometry modify N and P conservation strategies of a desert steppe species Glycyrrhiza 497 uralensis. Sci. Rep. 8, 12668. https://doi.org/10.1038/s41598-018-30324-w. 498 Huang, Z.Q., Liu, B., Davis, M., Sardans, J., Peñuelas, J., Billings, S., 2016. Long-term nitrogen 499 deposition linked to reduced water use efficiency in forests with low phosphorus availability. 500 New Phytol. 210, 431-442. https://doi.org/10.1111/nph.13785. 501 Jian, S.Y., Li, J.W., Chen, J., Wang, G.S., Mayes, M.A., Dzantor, K.E., Hui, D.F., Luo, Y.Q., 2016. 502 Soil extracellular enzyme activities, soil carbon and nitrogen storage under nitrogen fertilization: 503 a meta-analysis. Soil Biol. Biochem. 101, 32-43. https://doi.org/10.1016/j.soilbio.2016.07.003. 504 Kandeler, E., Gerber, H., 1988. Short-term assay of soil urease activity using colorimetric 505 determination of ammonium. Biol. Fertil. Soils 6, 68-72. https://doi.org/10.1007/BF00257924. 506 Keeler, B.L., Hobbie, S.E., Kellogg, L.E., 2009. Effects of long-term nitrogen addition on microbial 507 enzyme activity in eight forested and grassland sites: implications for litter and soil organic matter 508 decomposition. Ecosystems 12, 1-15. https://doi.org/10.1007/s10021-008-9199-z. 509 Lal, R., 2004. Soil carbon sequestration to mitigate climate change. Geoderma 123, 1-22. 510 https://doi.org/10.1016/j.geoderma.2004.01.032. 511 Lin, H.J., Hu, T.X., 2011. Preliminary study on the chemistry features of atmospheric precipitation of 512 three typical regions in Sichuan (in Chinese with English abstract), Sichuan Agricultural 513 University, Ya'an. 514 Lu, X., Gilliam, F.S., Yu, G., Li, L., Mao, Q., Chen, H., Mo, J., 2013. Long-term nitrogen addition 515 decreases carbon leaching in a nitrogen-rich forest ecosystem. Biogeosciences 10, 3931-3941. 516 https://doi.org/10.5194/bg-10-3931-2013. 517 Lv, Y., Wang, C.Y., Jia, Y.Y., Wang, W.W., Ma, X., Du, J.J., Pu, G.Z., Tian, X.J., 2014. Effects of 24 518 sulfuric, nitric, and mixed acid rain on litter decomposition, soil microbial biomass, and enzyme 519 activities in subtropical forests of China. Appl. Soil Ecol. 79, 1-9. 520 https://doi.org/10.1016/j.apsoil.2013.12.002. 521 Magill, A.H., Aber, J.D., 1998. Long-term effects of experimental nitrogen additions on foliar litter 522 decay and humus formation in forest ecosystems. Plant Soil 203, 301-311. 523 https://doi.org/10.1023/a:1004367000041. 524 Magill, A.H., Aber, J.D., Currie, W.S., Nadelhoffer, K.J., Martin, M.E., Mcdowell, W.H., Melillo, J.M., 525 Steudler, P., 2004. Ecosystem response to 15 years of chronic nitrogen additions at the Harvard 526 Forest LTER, Massachusetts, USA. For. Ecol. Manag. 196, 7-28. 527 https://doi.org/10.1016/j.foreco.2004.03.033. 528 Marklein, A.R., Houlton, B.Z., 2012. Nitrogen inputs accelerate phosphorus cycling rates across a 529 wide variety of terrestrial ecosystems. New Phytol. 193, 696-704. https://doi.org/10.1111/j.1469- 530 8137.2011.03967.x. 531 Matejko, M., Dore, A.J., Hall, J., Dore, C.J., Blas, M., Kryza, M., Smith, R., Fowler, D., 2009. The 532 influence of long term trends in pollutant emissions on deposition of sulphur and nitrogen and 533 exceedance of critical loads in the United Kingdom. Environ. Sci. Policy 12, 882-896. 534 https://doi.org/10.1016/j.envsci.2009.08.005. 535 Michopoulos, P., Baloutsos, G., Economou, A., Nikolis, N., 2004. Effects of nitrogen deposition on 536 nitrogen cycling in an Aleppo pine stand in Athens, Greece. Sci. Total Environ. 323, 211-218. 537 https://doi.org/10.1016/j.scitotenv.2003.10.002. 538 Mori, T., Zhou, K.J., Wang, S.H., Zhang, W., Mo, J.M., 2019. Effect of nitrogen addition on DOC 539 leaching and chemical exchanges on canopy leaves in Guangdong Province, China. J. For. Res. 540 30, 1707-1713. https://doi.org/10.1007/s11676-018-0800-9. 541 Ohshima, T., Tamura, T., Sato, M., 2007. Influence of pulsed electric field on various enzyme activities. 542 J. Electrostatics 65, 156-161. https://doi.org/10.1016/j.elstat.2006.07.005. 543 Oulehle, F., Chuman, T., Majer, V., Hruska, J., 2013. Chemical recovery of acidified Bohemian lakes 544 between 1984 and 2012: the role of acid deposition and bark beetle induced forest disturbance. 545 Biogeochemistry 116, 83-101. https://doi.org/10.1007/s10533-013-9865-x. 546 Paytan, A., McLaughlin, K., 2007. The oceanic phosphorus cycle. Chem. Rev. 107, 563-576. 547 https://doi.org/10.1021/cr0503613. 25 548 Peng, Y., Song, S.Y., Li, Z.Y., Li, S., Chen, G.T., Hu, H.L., Xie, J.L., Chen, G., Xiao, Y.L., Liu, L., 549 Tang, Y., Tu, L.H., 2020. Influences of of nitrogen addition and aboveground litter-input 550 manipulations on soil respiration and biochemical properties in a subtropical forest. Soil Biol. 551 Biochem. 142, 107694. https://doi.org/10.1016/j.soilbio.2019.107694. 552 Peng, Y.F., Peng, Z.P., Zeng, X.T., Houx, J.H., 2019. Effects of nitrogen-phosphorus imbalance on 553 plant biomass production: a global perspective. Plant Soil 436, 245-252. 554 https://doi.org/10.1007/s11104-018-03927-5. 555 Peñuelas, J., Poulter, B., Sardans, J., Ciais, P., Velde, M.V.D., Bopp, L., Boucher, O., Godderis, Y., 556 Hinsinger, P., Llusia, J., 2013. Human-induced nitrogen–phosphorus imbalances alter natural and 557 managed ecosystems across the globe. Nat. Commun. 4, 2934. 558 https://doi.org/10.1038/ncomms3934. 559 Prescott, C.E., 2010. Litter decomposition: what controls it and how can we alter it to sequester more 560 carbon in forest soils? Biogeochemistry 101, 133-149. https://doi.org/10.1007/s10533-010-9439- 561 0. 562 Saiya-Cork, K.R., Sinsabaugh, R.L., Zak, D.R., 2002. The effects of long term nitrogen deposition on 563 extracellular enzyme activity in an Acer saccharum forest soil. Soil Biol. Biochem. 34, 1309- 564 1315. https://doi.org/10.1016/s0038-0717(02)00074-3. 565 Sardans, J., Peñuelas, J., 2012. The role of plants in the effects of global change on nutrient availability 566 and stoichiometry in the plant-soil system. Plant Physiol. 160, 1741-1761. 567 https://doi.org/10.1104/pp.112.208785. 568 Sardans, J., Peñuelas, J., 2015. Trees increase their P:N ratio with size. Glob. Ecol. Biogeogr. 24, 147- 569 156. https://doi.org/10.1111/geb.12231. 570 Schinner, F., Ohlinger, R., Kandeler, E., Margesin, R., 1996. Methods in soil biology. Springer-Verlag, 571 Berlin. 572 Scott, J.T., Lambie, S.M., Stevenson, B.A., Schipper, L.A., Parfitt, R.L., McGill, A.C., 2015. Carbon 573 and nitrogen leaching under high and low phosphate fertility pasture with increasing nitrogen 574 inputs. Agric. Ecosyst. Environ. 202, 139-147. https://doi.org/10.1016/j.agee.2014.12.021. 575 Tabatabai, M.A., 1982. Soil enzymes. In: Weaver, R., Angle, S., Bottomley, P., Bezdicek, D., Smith, 576 S., Tabatabai, M., Wollum, A. (Eds.), Methods of Soil Analysis. Soil Science Society of America, 577 Madison, WI, USA, pp. 903-947. 26 578 Tan, J., Fu, J.S., Dentener, F., Jian, S., Keating, T., 2018. Multi-model study of HTAP II on sulfur and 579 nitrogen deposition. Atmos. Chem. Phys. 18, 6847-6866. https://doi.org/10.5194/acp-18-6847- 580 2018. 581 Tie, L.H., Fu, R., Zhang, S.B., Zhou, S.X., Han, B.H., Huang, C.d., 2018. Effects of simulated nitrogen 582 and sulfur deposition on litter decomposition rate in an evergreen broad-leaved forest in the Rainy 583 Area of Western China (in Chinese with English abstract). J. Appl. Ecol. 29, 2243-2250. 584 https://doi.org/10.13287/j.1001-9332.201807.012. 585 Tu, L.H., Hu, T.X., Zhang, J., Li, X.W., Hu, H.L., Liu, L., Xiao, Y.L., 2013. Nitrogen addition 586 stimulates different components of soil respiration in a subtropical bamboo ecosystem. Soil Biol. 587 Biochem. 58, 255-264. https://doi.org/10.1016/j.soilbio.2012.12.005. 588 Turner, B.L., 2008. Resource partitioning for soil phosphorus: a hypothesis. J. Ecol. 96, 698-702. 589 https://doi.org/10.1111/j.1365-2745.2008.01384.x. 590 van Dijk, J., Robroek, B., Kardel, I., Wassen, M., 2012. Combined effects of nitrogen enrichment, 591 sulphur pollution and climate change on fen meadow vegetation N:P stoichiometry and biomass. 592 Biogeochemistry 111, 139-150. https://doi.org/10.1007/s10533-011-9694-8. 593 Xu, Y.H., Fan, J.L., Ding, W.X., Gunina, A., Chen, Z.M., Bol, R., Luo, J.F., Bolan, N., 2017. 594 Characterization of organic carbon in decomposing litter exposed to nitrogen and sulfur additions: 595 links to microbial community composition and activity. Geoderma 286, 116-124. 596 https://doi.org/10.1016/j.geoderma.2016.10.032. 597 Xu, Z., Tu, L., Hu, T., Schädler, M., 2013. Implications of greater than average increases in nitrogen 598 deposition on the western edge of the Szechwan Basin, China. Environ. Pollut. 177, 201-201. 599 https://doi.org/10.1016/j.envpol.2012.12.031. 600 Yuan, Z.Y., Chen, H.Y.H., 2015. Decoupling of nitrogen and phosphorus in terrestrial plants associated 601 with global changes. Nat. Clim. Chang. 5, 465-469. https://doi.org/10.1038/nclimate2549. 602 Zechmeister-Boltenstern, S., Keiblinger, K.M., Mooshammer, M., Peñuelas, J., Richter, A., Sardans, 603 J., Wanek, W., 2015. The application of ecological stoichiometry to plant-microbial-soil organic 604 matter transformations. Ecol. Monogr. 85, 133-155. https://doi.org/10.1890/14-0777.1. 605 Zhang, T.A., Chen, H.Y.H., Ruan, H.H., 2018. Global negative effects of nitrogen deposition on soil 606 microbes. ISME J. 12, 1817-1825. https://doi.org/10.1038/s41396-018-0096-y. 607 Zhou, S.X., Xiang, Y.B., Tie, L.H., Han, B.H., Huang, C.D., 2018. Simulated nitrogen deposition 27 608 significantly reduces soil respiration in an evergreen broadleaf forest in western China. PLoS 609 ONE 13, e0204661. https://doi.org/10.1371/journal.pone.0204661. 610 28 611 Tables 612 Table 1 F- and P-values of linear mixed effects models on the effects of nitrogen (N) addition, sulfur 613 (S) addition, sampling time (month) and their interactions on the concentrations of SOC, TN, and TP 614 and the ratios of C/N, C/P, and N/P. d.f., degrees of freedom; SOC, soil organic carbon; TN, total 615 nitrogen; TP, total phosphorus; C/N, ratio of SOC to TN concentrations; C/P, ratio of SOC to TP 616 concentrations; and N/P, ratio of TN to TP concentrations. n = 576. d.f. F-value P-value d.f. F-value P-value -1 SOC (g kg ) C/N ratio N 1 66.7 <0.001 N 1 564 <0.001 S 1 17.6 <0.010 S 1 363 <0.001 Time 7 2265 <0.001 Time 7 2170 <0.001 N×S 1 5.70 <0.050 N×S 1 28.9 <0.001 N × Time 7 118 <0.001 N × Time 7 124 <0.001 S × Time 7 26.3 <0.001 S × Time 7 194 <0.001 N × S × Time 7 85.4 <0.001 N × S × Time 7 159 <0.001 TN (g kg-1) C/P ratio N 1 32.1 <0.001 N 1 40.7 <0.001 S 1 30.4 <0.001 S 1 1.65 0.223 Time 7 682 <0.001 Time 7 439 <0.001 N×S 1 0.352 0.564 N×S 1 0.600 0.454 N × Time 7 118 <0.001 N × Time 7 30.8 <0.001 S × Time 7 124 <0.001 S × Time 7 13.1 <0.001 N × S × Time 7 94.7 <0.001 N × S × Time 7 30.4 <0.001 -1 TP (g kg ) N/P ratio N 1 0.001 0.975 N 1 57.1 <0.001 S 1 1.16 0.303 S 1 91.1 <0.001 Time 7 80.6 <0.001 Time 7 219 <0.001 N×S 1 0.307 0.590 N×S 1 0.009 0.926 N × Time 7 11.0 <0.001 N × Time 7 18.3 <0.001 S × Time 7 16.8 <0.001 S × Time 7 39.5 <0.001 N × S × Time 7 8.49 <0.001 N × S × Time 7 12.1 <0.001 617 29 618 Table 2 F- and P-values of linear mixed effects models on the effects of nitrogen (N) addition, sulfur 619 (S) addition, sampling time (month) and their interactions on the activities of soil invertase, urease, 620 acid phosphatase, catalase, cellulase, peroxidase, and polyphenol oxidase. d.f., degrees of freedom. n 621 = 672. d.f. F-value P-value d.f. F-value P-value Invertase (μmol g-1 h-1) Catalase (μmol g-1 h-1) N 1 90.7 <0.001 N 1 18.4 <0.010 S 1 379 <0.001 S 1 1.12 0.310 Time 7 119 <0.001 Time 7 458 <0.001 N×S 1 0.610 0.501 N×S 1 5.95 <0.050 N × Time 7 8.80 <0.001 N × Time 7 23.4 <0.001 S × Time 7 15.9 <0.001 S × Time 7 64.4 <0.001 N × S × Time 7 4.07 <0.010 N × S × Time 7 17.5 <0.001 -1 -1 -1 -1 Urease (μmol g h ) Peroxidase (μmol g h ) N 1 41.8 <0.001 N 1 12.2 <0.010 S 1 6.42 <0.050 S 1 4.38 0.058 Time 7 553 <0.001 Time 7 652 <0.001 N×S 1 3.21 0.098 N×S 1 1.30 0.277 N × Time 7 80.8 <0.001 N × Time 7 99.1 <0.001 S × Time 7 8.98 <0.001 S × Time 7 101 <0.001 N × S × Time 7 12.7 <0.001 N × S × Time 7 34.1 <0.001 -1 -1 -1 -1 Acid phosphatase (μmol g h ) Polyphenol oxidase (μmol g h ) N 1 4.51 <0.050 N 1 15.6 <0.010 S 1 5.09 <0.050 S 1 23.4 <0.001 Time 7 122 <0.001 Time 7 7277 <0.001 N×S 1 0.963 0.346 N×S 1 0.205 0.659 N × Time 7 14.6 <0.001 N × Time 7 20.3 <0.001 S × Time 7 12.6 <0.001 S × Time 7 33.2 <0.001 N × S × Time 7 9.24 <0.001 N × S × Time 7 80.0 <0.001 -1 -1 Cellulase (μmol g h ) N 1 37.1 <0.001 S 1 0.014 0.907 Time 7 81.9 <0.001 N×S 1 16.3 <0.010 N × Time 7 23.6 <0.001 S × Time 7 27.1 <0.001 N × S × Time 7 42.3 <0.001 622 30 623 Table 3 Standardized total effects, direct effects, and indirect effects of nitrogen (N) and sulfur (S) 624 additions on soil organic carbon (SOC), total nitrogen (TN), and total phosphorus (TP) in the structural 625 equation models. **: highly significant (P < 0.01); *: significant (P < 0.05). n = 1088. Relationships Direct effects Indirect effects Total effects N addition on SOC 0.214* -0.582** -0.368** S addition on SOC 0.157* -0.346** -0.189** N addition on TN 0.147* 0.317** 0.464** S addition on TN 0.210** 0.241** 0.451** N addition on TP No significant effects 0.0022* 0.0022* S addition on TP No significant effects -0.157* -0.157* 626 31 627 628 Figure captions 629 630 Fig. 1. Effects of nitrogen and sulfur additions on the mean concentrations of SOC (a), TN (b), and TP 631 (c), and mean ratios of C/N (d), C/P (e), and N/P (f) during two years of fertilization based on linear 632 mixed effects models (mean ± standard deviations). Different lowercase letters denote significant 633 differences (multiple comparisons with Bonferroni adjustment, P < 0.05) among different treatments 634 during two years of fertilization. SOC: soil organic carbon; TN: soil total nitrogen; TP: soil total 635 phosphorus; C/N: ratio of SOC to TN concentrations; C/P: ratio of SOC to TP concentrations; N/P: 636 ratio of TN to TP concentrations; Ctr: no added nitrogen or sulfur; +N: added nitrogen; +S: added 637 sulfur; and +NS: added nitrogen and sulfur. n = 576. 638 639 Fig. 2. Effects of nitrogen and sulfur additions on the mean activities of soil invertase (a), urease (b), 640 acid phosphatase (c), cellulase (d), catalase (e), peroxidase (f), and polyphenol oxidase (g) during two 641 years of fertilization based on linear mixed effects models (mean ± standard deviations). Different 642 lowercase letters denote significant differences (multiple comparisons with Bonferroni adjustment, P 643 < 0.05) among different treatments during two years of fertilization. Ctr: no added nitrogen or sulfur; 644 +N: added nitrogen; +S: added sulfur; and +NS: added nitrogen and sulfur. n = 672. 645 646 Fig. 3. Relationships between SOC and TN (a), SOC and N/P (b), SOC and TP (c), TN and C/P (d), 647 TP and TN (e), and TP and C/N (f) in the treatments during two years of fertilization. **: highly 648 significant (P < 0.01) correlation; *: significant (P < 0.05) correlation. The solid regression lines 649 denote significant (P < 0.05) or highly significant (P < 0.01) correlations. SOC: soil organic carbon; 650 TN: soil total nitrogen; TP: soil total phosphorus; C/N: ratio of SOC to TN concentrations; C/P: ratio 651 of SOC to TP concentrations; N/P: ratio of TN to TP concentrations; Ctr: no added nitrogen or sulfur; 652 +N: added nitrogen; +S: added sulfur; and +NS: added nitrogen and sulfur. n = 576. 653 654 Fig. 4. Structural equation models of the effects of nitrogen (N) and sulfur (S) additions on the soil 655 enzyme activities and the relative contributions of the activities to the SOC (a, b), TN (c, d), and TP 656 (e, f) concentrations during two years of fertilization. Solid black and dashed blue arrows denote 32 657 significant (P < 0.05) positive and negative effects, respectively. SOC, soil organic carbon; TN, soil 658 total nitrogen; TP, soil total phosphorus; INV, invertase; URE, urease; AP, acid phosphatase; CA, 659 catalase; CEL, cellulase; and POD, peroxidase. **, highly significant (P < 0.01); and *, significant (P 660 < 0.05). n = 1088. 661 662 Fig. 5. Two-dimensional figure formed by the two first roots of the general discriminant analysis 663 (GDA) showing the area formed by the mean ± confidence interval (95%) of the scores of the control 664 and three treatment cases. The loads in these two roots of the independent continuous variables SOC, 665 TN and TP concentrations; C/N, C/P and N/P ratios; and soil enzyme activities are also represented. 666 SOC: soil organic carbon; TN: soil total nitrogen; TP: soil total phosphorus; C/N: ratio of SOC to TN 667 concentrations; C/P: ratio of SOC to TP concentrations; N/P: ratio of TN to TP concentrations; Ctr: no 668 added nitrogen or sulfur; +N: added nitrogen; +S: added sulfur; and +NS: added nitrogen and sulfur. 669 33 670 Figures 671 672 Fig. 1 16 a (a) 2.5 (b) bc b a c 2.0 b b 12 c TN (g kg-1) SOC (g kg-1) 1.5 8 1.0 4 0.5 0 0.0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 0.6 a a a (c) 12 a (d) a 10 b b TP (g kg-1) C/N ratio 0.4 8 c 6 0.2 4 2 0.0 0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 35 a (e) 5.0 a (f) ab b 30 b b 4.0 b 25 c C/P ratio N/P ratio 20 3.0 15 2.0 10 5 1.0 0 0.0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 673 34 674 Fig. 2 4.0 (a) 5.0 (b) a a a b (μmol g-1 h-1) b (μmol g-1 h-1) 3.0 c d 4.0 c Invertase Urease 3.0 2.0 2.0 1.0 1.0 0.0 0.0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 2.0 (c) 2.0 (d) a a a a b Acid phosphatase b b c (μmol g-1 h-1) (μmol g-1 h-1) 1.5 1.5 Cellulase 1.0 1.0 0.5 0.5 0.0 0.0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 0.4 (e) 50 (f) a b a bc b ab b (μmol g-1 h-1) c 40 (μmol g-1 h-1) 0.3 Peroxidase Catalase 30 0.2 20 0.1 10 0.0 0 Ctr +N +S +NS Ctr +N +S +NS Treatment Treatment 10 (g) a Polyphenol oxidase b bc c 8 (μmol g-1 h-1) 6 4 2 0 Ctr +N +S +NS Treatment 675 35 676 Fig. 3 677 Ctr +N +S +NS Ctr +N +S +NS Ctr +N +S +NS Ctr +N +S +NS 18 18 (a) (b) 16 16 Ctr: R2 = 0.33* 14 14 +N: R2 = 0.51** SOC (g kg-1) SOC (g kg-1) +S: R2 = 0.26* 12 12 +NS: R2 = 0.54** Ctr: R2 = 0.30* 10 +N: R2 = 0.46** 10 8 +S: R2 = 0.21* 8 +NS: R2 = 0.60** 6 6 0.8 1.2 1.6 2 2.4 1 2 3 4 5 6 7 TN (g kg-1) N/P ratio Ctr +N +S +NS +NS Ctr +N +S +NS +N +S +NS 18 2.5 (c) (d) 16 2.0 14 SOC (g kg-1) TN (g kg-1) 12 1.5 Ctr: R2 = 0.06 Ctr: R2 = 0.04 +N: R2 = 0.48** 10 +N: R2 = 0.09 1.0 +S: R2 = 0.18* 8 +S: R2 = 0.08 +NS: R2 = 0.38** +NS: R2 = 0.20** 6 0.5 0.3 0.4 0.5 0.6 0.7 10 15 20 25 30 35 40 TP (g kg-1) C/P ratio Ctr +N +S +NS Ctr Ctr +N +S +NS Ctr +NS 0.6 0.6 (e) (f) 0.5 0.5 TP (g kg-1) TP (g kg-1) 0.4 0.4 Ctr: R2 = 0.17* Ctr: R2 = 0.21* +N: R2 = 0.00 +N: R2 = 0.04 0.3 +S: R2 = 0.00 0.3 +S: R2 = 0.05 +NS: R2 = 0.12 +NS: R2 = 0.18* 0.2 0.2 0.8 1.2 1.6 2 2.4 2 4 6 8 10 12 14 16 TN (g kg-1) C/N ratio 678 36 679 Fig. 4 N S (a) 0.30** -0.35** -0.24* 0.57** (b) -0.22* -0.21* -0.33** 0.37** R2=0.33** -0.48** R2=0.16* R2=0.11** R2=0.11** R2=0.12** R2=0.23** R2=0.15* R2=0.23** R2=0.14* 0.16* INV URE CEL AP POD INV URE AP POD 0.21* 0.36** 0.17* -0.15** -0.20** -0.24** -0.16** 0.77** 0.34* 0.46** SOC SOC R2=0.69** R2=0.53** N S (c) 0.30** -0.35** (d) 0.37** -0.21** 0.57** -0.22** R2=0.33** R2=0.11** R2=0.12** R2=0.15* R2=0.23** R2=0.14** URE AP POD URE AP POD 0.15* 0.21** 0.62** 0.62** 0.11* -0.40** 0.11* -0.40** TN TN R2=0.72** R2=0.73** N S (e) 0.30* -0.35** (f) -0.21* -0.40** 0.30** 0.57** -0.22* R2=0.33** R2=0.11* R2=0.16** R2=0.15* R2=0.23** R2=0.12* R2=0.14* URE AP CA POD URE AP POD 0.14* 0.42** 0.14* 0.54** 0.52** 0.35** 0.63** TP TP R2=0.67** R2=0.56** 680 37 681 Fig. 5 Loadings -0.2 -0.1 0 0.1 0.2 3 +N Urease +NS 0.3 TN 2 0.2 N/P A. Phosphatase Root 2 (28.9%) 1 0.1 Scores Loadings TP 0 0 Invertase Polyphenol Oxidase -1 -0.1 Control Catalase Peroxidase Cellulase -2 C/N SOC -0.2 C/P -3 +S -0.3 -5 -2.5 0 2.5 5 Root 1 (51.7%) Scores 682 38 683 Supplementary tables and figures 684 Table S1 Enzymes used in this study and their measurements Enzyme Preparation of enzyme solution Incubation condition Measurement of activity Invertase 5 mL phosphate buffer (pH 5.5) 37 °C, 24 h Production rate of glucose from sucrose Urease 10 mL citrate buffer (pH 6.7) 37 °C, 24 h Production rate of ammonia from urea Production rate of phenol from para-nitrophenyl Acid phosphatase 20 mL acetate buffer (pH 5.0) 37 °C, 2 h phosphate Production rate of glucose from carboxymethylcellulose Cellulase 10 mL acetate buffer (pH 5.5) 37 °C, 72 h sodium Catalase 40 mL distilled water 25 °C, 0.5 h Degradation rate of hydrogen peroxide 4 mL citrate-phosphate buffer Production rate of phenanthraquinone from catechol in Peroxidase 30 °C, 2 h (pH 4.5) the presence of H2O2 4 mL citrate-phosphate buffer Production rate of purple gallopoietin from pyrogallic Polyphenol oxidase 30 °C, 2 h (pH 4.5) acid without H2O2 685 39 686 Table S2 Squared Mahalanobis distances (SMDs) and F- and P-values in the general discriminant 687 analysis (GDA) among the treatments and control, with SOC, TN and TP concentrations; C/N, C/P 688 and N/P ratios; and soil enzyme activities as the continuous independent variables and time of 689 sampling (month) as the categorical independent controlling variable. SOC: soil organic carbon; TN: 690 soil total nitrogen; TP: soil total phosphorus; C/N: ratio of SOC to TN concentrations; C/P: ratio of 691 SOC to TP concentrations; N/P: ratio of TN to TP concentrations; Ctr: no added nitrogen or sulfur; 692 +N: added nitrogen; +S: added sulfur; and +NS: added nitrogen and sulfur. +N +S +NS Treatment SMD F-value P-value SMD F-value P-value SMD F-value P-value Ctr 36.6 17.4 <0.001 44.4 21.2 <0.001 64.1 30.5 <0.001 +N 58.8 28.0 <0.001 66.3 31.6 <0.001 +S 41.6 19.8 <0.001 693 40 694 Table S3 Main effects of the variables in the general discriminant analysis (GDA). Statistics (Wilks’ 695 λ and F- and P-values) of the discriminant functional analysis among the treatments and control, 696 with SOC, TN and TP concentrations; C/N, C/P, and N/P ratios; and soil enzyme activities as the 697 independent continuous variables and time of sampling as the independent controlling categorical 698 variable. The significant effects of a variable in the model are highlighted in bold type (P < 0.05). 699 SOC: soil organic carbon; TN: total nitrogen; TP: total phosphorus; C/N: ratio of SOC to TN 700 concentrations; C/P: ratio of SOC to TP concentrations; N/P: ratio of TN to TP concentrations. Variables Wilks’ Lambda F-value P-value SOC (g kg-1) 0.962 0.972 0.410 -1 TN (g kg ) 0.792 6.37 <0.001 TP (g kg-1) 0.990 0.240 0.868 C/N 0.702 10.3 <0.001 C/P 0.962 0.964 0.410 N/P 0.994 0.149 0.930 Invertase (µmol g-1 h-1) 0.749 8.16 <0.001 -1 -1 Urease (µmol g h ) 0.276 63.8 <0.001 -1 -1 Cellulase (µmol g h ) 0.792 6.38 <0.001 Acid phosphatase (µmol g-1 h-1) 0.409 35.2 <0.001 Catalase (µmol g-1 h-1) 0.604 15.9 <0.001 -1 1 Peroxidase (µmol g h- ) 0.831 4.94 <0.010 -1 -1 Polyphenol oxidase (µmol g h ) 0.351 45.1 <0.001 Time 0.060 16.7 <0.001 701 41 702 Fig. S1 Dynamics of soil-water content and soil temperature in the 0-20 cm soil layer from April 703 2013 to April 2015 (mean ± standard deviation). 40 40 Soil temperature (℃) Soil temperature Soil-water content Soil-water content (%) 30 30 20 20 10 10 0 0 Time (month) 704 42 705 Fig. S2 Changes in the SOC (a), TN (b), and TP (c) concentrations in the treatments at each sampling 706 time based on a two-way ANOVA (mean ± standard deviations). Different lowercase letters denote 707 significant differences (multiple comparisons with Bonferroni adjustment, P < 0.05) among different 708 treatments at each sampling time. SOC, soil organic carbon; TN, soil total nitrogen; TP, soil total 709 phosphorus; Ctr, no added nitrogen or sulfur; +N, added nitrogen; +S, added sulfur; and +NS, added 710 nitrogen and sulfur. n = 288. 20 (a) a Ctr +N +S +NS aaaa a abb b a SOC (g kg-1) c a 15 c bd a b a c c bbb a a c bc cbc 10 d 5 0 3 6 9 12 15 18 21 24 Time (month) 3.0 (b) Ctr +N +S +NS a a a a a abb a bb TN (g kg-1) b bbb bb 2.0 aaaa a cbc c b c c cc b c 1.0 0.0 3 6 9 12 15 18 21 24 Time (month) 0.8 (c) Ctr +N +S +NS aa a a 0.6 aaaa a a a a a a a a a a a a a aa a a a a a TP (g kg-1) a aa a 0.4 0.2 0.0 3 6 9 12 15 18 21 24 Time (month) 711 712 43 713 Fig. S3 Effects of nitrogen and sulfur additions on the mean soil pH value during two years of 714 fertilization based on linear mixed effects models (mean ± standard deviation). Different lowercase 715 letters denote significant differences (multiple comparisons with Bonferroni adjustment, P < 0.05) 716 among different treatments during two years of fertilization. Ctr: no added nitrogen or sulfur; +N: 717 added nitrogen; +S: added sulfur; and +NS: added nitrogen and sulfur. n = 96. 4.6 Soil pH (H2O extraction) 4.5 a b 4.4 b 4.3 c 4.2 4.1 4.0 Ctr +N +S +NS Treatment 718 719 44 720 Fig. S4 Changes in the C/N (a), C/P (b), and N/P (c) ratios in the treatments at each sampling time 721 based on a two-way ANOVA (mean ± standard deviations). Different lowercase letters denote 722 significant differences (multiple comparisons with Bonferroni adjustment, P < 0.05) among different 723 treatments at each sampling time. C/N, ratio of soil organic carbon to total nitrogen concentrations; 724 C/P, ratio of soil organic carbon to total phosphorus concentrations; N/P, ratio of soil total nitrogen to 725 total phosphorus concentrations; Ctr, no added nitrogen or sulfur; +N, added nitrogen; +S, added sulfur; 726 and +NS, added nitrogen and sulfur. n = 288. 20 (a) Ctr +N +S +NS a 15 b a C/N ratio aba a b c a d a a a a a 10 b b c cbc bb c a c cbc bb c 5 0 3 6 9 12 15 18 21 24 Time (month) 60 (b) Ctr +N +S +NS aaaa a C/P ratio 40 ab abaab a ab b a bc c a aab a a bb bc c c c b c bb b 20 0 3 6 9 12 15 18 21 24 Time (month) 10 (c) Ctr +N +S +NS a N/P ratio a ba a bb 5 a bb a a a bbb a c b b ba a a c c bb c cb d 0 3 6 9 12 15 18 21 24 Time (month) 727 728 45 729 Fig. S5 Changes in the activities of soil invertase (a), urease (b), acid phosphatase (c), cellulase (d), 730 catalase (e), peroxidase (f), and polyphenol oxidase (g) in the treatments at each sampling time based 731 on a two-way ANOVA (mean ± standard deviations). Different lowercase letters denote significant 732 differences (multiple comparisons with Bonferroni adjustment, P < 0.05) among different treatments 733 at each sampling time. Ctr, no added nitrogen or sulfur, +N, added nitrogen; +S, added sulfur; and 734 +NS, added nitrogen and sulfur. n = 672. 4.0 (a) Ctr +N +S +NS 6.0 (b) Ctr +N +S +NS aa aab a a a a a 3.0 a b a a b aaa bb bb a a 5.0 abab a ab a ab a bab a a a a ab ab a ab c b c b (μmol g-1 h-1) (μmol g-1 h-1) b b b cd bc 4.0 a a b a b b b b c c Invertase c Urease 2.0 d 3.0 2.0 1.0 1.0 0.0 0.0 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 Time (month) Time (month) 2.5 (c) Ctr +N +S +NS 2.5 (d) Ctr +N +S +NS a a a a a ababa aaaa a a a Acid phosphatase (μmol g-1 h-1) a aba 2.0 aaaa a aaba a a a aa a a a a a ab b 2.0 b b bbb bb a bb Cellulase (μmol g-1 h-1) b b aa bb b 1.5 b 1.5 c c c c bb bb 1.0 1.0 0.5 0.5 0.0 0.0 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 Time (month) Time (month) 0.4 a aa (e) a aa Ctr +N +S +NS 60 (f) Ctr+N +S +NS a a ab a ab a a a a a a (μmol g-1 h-1) (μmol g-1 h-1) a a b bc b ab a ab b bb b ab b b b b b Peroxidase 0.3 c a a a a aaa a a a abc bc a a a b b b a a a Catalase b 40 a a bb b 0.2 20 0.1 0.0 0 3 6 9 12 15 18 21 24 3 6 9 12 15 18 21 24 Time (month) Time (month) 12 (g) Ctr +N +S +NS a aa aaa b Polyphenol oxidase a b 9 a a ab a aabb b a b b b bbb (μmol g-1 h-1) a bb bbb b 6 3 0 3 6 9 12 15 18 21 24 Time (month) 735 736 737 46