This is the accepted version of the article: Wang, Weiqi; Sardans i Galobart, Jordi; Wang, Chun; [et al.]. The response of stocks of C, N, and P to plant invasion in the coastal wetlands of China. DOI 10.1111/gcb.14491 This version is avaible at https://ddd.uab.cat/record/203307 under the terms of the license 1 The response of stocks of C, N and P to plant invasion in 2 the coastal wetlands of China 3 Weiqi Wang1,2,3,4*, Jordi Sardans3,4*, Chun Wang1,2, Congsheng Zeng1,2, Chuan Tong1,2, 4 Guixiang Chen1,2, Jiafang Huang1,2, Haoran Pan5, Guille Peguero3,4, Helena Vallicrosa3,4, 5 Josep Peñuelas3,4 6 1 7 Key Laboratory of Humid Subtropical Eco-geographical Process, Ministry of Education, 8 Fujian Normal University, Fuzhou 350007, China 2 9 Institute of Geography, Fujian Normal University, Fuzhou 350007, China 3 10 CSIC, Global Ecology Unit CREAF-CSIC-UAB, 08913 Cerdanyola del Vallès, Catalonia, 11 Spain 4 12 CREAF, 08913 Cerdanyola del Vallès, Catalonia, Spain 5 13 Guangxi Mangrove Research Center, Guangxi Academy of Sceinces, Beihai 536000, 14 China 15 16 Correspondence 17 Weiqi Wang, Key Laboratory of Humid Subtropical Eco-geographical Process, Ministry of 18 Education, Fujian Normal University, Fuzhou 350007, China, Institute of Geography, 19 Fujian Normal University, Fuzhou 350007, China. Email: wangweiqi15@163.com; 20 Jordi Sardans, CSIC, Global Ecology Unit CREAF-CSIC-UAB, 08913 Cerdanyola del 21 Vallès, Catalonia, Spain, CREAF, 08913 Cerdanyola del Vallès, Catalonia, Spain. Email: 22 j.sardans@creaf.uab.es 23 1 24 Abstract 25 The increasing success of invasive plant species in wetland areas can threaten their capacity 26 to store carbon, nitrogen, and phosphorus (C, N, and P). Here we have investigated the 27 relationships between the different stocks of soil organic carbon (SOC), and total C, N, and 28 P pools in the plant-soil system from eight different wetland areas across the South-East coast 29 of China, where the invasive tallgrass Spartina alterniflora has replaced the native tall 30 grasses Phragmites australis and the mangrove communities, originally dominated by the 31 native species Kandelia obovata and Avicennia marina. The invasive success of Spartina 32 alterniflora replacing Phragmites australis did not greatly influence soil traits, biomass 33 accumulation or plant-soil C and N storing capacity. However, the resulting higher ability to 34 store P in both soil and standing plant biomass (approximately more than 70 and 15 kg P by 35 ha, respectively) in the invasive than in the native tall grass communities suggesting the 36 possibility of a decrease in the ecosystem N:P ratio with future consequences to below- and 37 above-ground trophic chains. The results also showed that a future advance in the native 38 mangrove replacement by Spartina alterniflora could constitute a serious environmental 39 problem. This includes enrichment of sand in the soil, with the consequent loss of nutrient 40 retention capacity, as well as a sharp decrease of the stocks of C (2.6 and 2.2 t C ha-1 in soil 41 and stand biomass, respectively), N, and P in the plant-soil system. This should be associated 42 with a worsening of the water quality by aggravating potential eutrophication processes. 43 Moreover, the loss of carbon and nutrient decreases the potential overall fertility of the system, 44 strongly hampering the re-establishment of woody mangrove communities in the future. 45 46 KEYWORDS 47 Active carbon, plant invasion, soil organic carbon, nutrient stoichiometry 2 48 INTRODUCTION 49 The world contains an estimated 2011 Pg organic carbon of soil (IPCC, 2000). The 50 importance of soil as a carbon sink is, therefore, crucial for the ability of the Earth to buffer 51 the increasing levels of atmospheric CO2 as a consequence of human activities (Joiner et al., 52 1999; Smith, 2004; Bonan & Van Cleve, 2011). Wetlands are one of the most sensitive 53 ecosystems to global climate change. Despite accounting just for 4 to 6% of the land area 54 (Howe et al., 2009), wetlands accumulate between 20 to 30% of the whole terrestrial soil C 55 stock (Smith et al., 2004), thus playing a disproportionate role in the global C cycle. As the 56 ecosystems with the world’s highest productivity per unit of area, wetlands combine the dual 57 function of being a “C source” and a “C sink” (Andreetta et al., 2016). Therefore, even minor 58 changes in the dynamics of wetland areas will affect global greenhouse gas emissions, which 59 in turn will influence global warming (Tian et al., 2010). Among wetland areas, estuarine 60 wetland soils, due to their location at the river-ocean interface, have a particularly large 61 potential to act as sediment sinks, thus accumulating C (Bianchi et al., 2013; Burden et al., 62 2013; Mitsch et al., 2013). In this context, the changes occurred in the last decades in China 63 coastal wetlands, such as the reduction in area because of sea level rise (Wang et al., 2015b) 64 and land use change (Wang et al., 2014) are especially important. Despite this reduction, 65 coastal wetlands in China currently still cover 5.80 × 104 km2 (Wetland China, 2014) and 66 provide many ecosystem services and products (Liu et al., 2006; Wang et al., 2015a). We 67 know that increases in tropical storms and ocean levels have altered flooding intensities, 68 which together with increased pollution and nutrient loads have impacted sedimentary 69 processes and the capacity of wetlands to store and release C and nutrients (Schewe et al., 70 2011; Mendelsohn et al., 2012; Sardans et al., 2012; Ramsar, 2013; Piecuch & Ponte, 2014; 3 71 Sardans & Peñuelas, 2014). In this context, the success of invasive plant species can also 72 have an impact on the capacity of wetlands to store and release C, N and P (Wang et al., 2014; 73 Wang et al., 2015b). 74 Invasive plant species can change the capacity to accumulate N and P in biomass and 75 soil (Wang et al., 2015a). The impacts of successful invasive plants on the concentrations and 76 stoichiometry of soil nutrients have been widely studied (Funk & Vitousek, 2007; Sardans & 77 Peñuelas, 2012), and some general trends have emerged (González et al., 2010; Matzek, 2011; 78 Sardans & Peñuelas, 2012; Sardans et al., 2017). The impacts of invasive plants can differ 79 and even be opposite depending on the natural availability levels of soil nutrients and, 80 generally, on the capacity of a site to sustain low or high plant production in accordance with 81 the soil and climatic conditions (González et al., 2010; Matzek, 2011; Sardans & Peñuelas, 82 2012; Sardans et al., 2017). 83 The active C from the total soil organic carbon, is highly susceptible to oxidation and 84 decomposition and is strongly influenced by plants and microorganisms (Kimura et al., 2004; 85 Chen et al., 2010) and includes dissolved organic carbon (DOC), labile organic carbon (LOC), 86 and soil microbial biomass organic carbon (MBC). Differently, the total content of soil 87 organic carbon (SOC) changes over a long-time scale, and their fluctuations are not easily 88 discerned within a short time period (Wissing et al., 2011). Therefore, distinguishing between 89 the active-C fraction from the total SOC pool is important to assess the effect of plant 90 invasion on soil C dynamics. Different forms of active SOC have different sensitivities to 91 environmental change (Gu et al., 2004), but few studies have examined the relationships 92 between the different forms of active SOC and changes in other environmental parameters 93 (Xu et al., 2011; Zhao et al., 2014), and especially of soil properties. Active SOC is a major 94 source of CO2 and CH4 produced by microbes, so properly managing this pool of active C is 4 95 important for mitigating global climate change (Knoblauch et al., 2011; Hanke et al., 2013). 96 This C:nutrient stoichiometry is a good indicator of changes in soil C dynamics in wetlands 97 due to changes in the environment, thereby providing information of the impacts on nutrient 98 cycling and status and, thus, is informative about the potential capacity of these ecosystems 99 to fix C and reduce the impact of the emissions of C-source greenhouse gases (Peñuelas et 100 al., 2012 and 2013). 101 The impacts of plant invasion on total soil C accumulation in the various soil fractions 102 and the further relationships of these impacts on C with other soil traits, such as C:N:P ratios, 103 are also unresolved (Sardans & Peñuelas, 2012). Most available data generally suggest that 104 the low costs of foliar construction and high phenotypic plasticity in taking up available 105 nutrients frequently contribute to invasive species success in nutrient-rich environments 106 (Daehler, 2003; González et al., 2010; Sardans & Peñuelas, 2012), whereas in nutrient-poor 107 soils the success of invasive plants would depend on more conservative strategies, such as a 108 higher nutrient-use efficiency (Funk & Vitousek, 2007; González et al., 2010; Matzek, 2011; 109 Sardans et al., 2017), especially on short timescales (Funk & Vitousek, 2007). The existence 110 of numerous exceptions, however, prevent this question from being totally resolved (Sardans 111 & Peñuelas, 2012). 112 Some studies have discussed soil C dynamics and nutrient stoichiometry (Tian et al., 113 2010; Schipper & Sparling, 2011), but few have distinguished them at a fine scale in soil 114 profiles. Studies addressing the relationships of species invasive success with soil C and 115 nutrient status has been mainly focused on the first centimeters of soil (Sardans & Peñuelas, 116 2012). However, there is little available information about the impact of plant invasive 117 success on soil depth. The different soil textures and soil organic matter decomposition 118 observed in superficial soil layers linked to plant alien success, can thereafter produce a shift 5 119 in the accumulation of some organic carbon fractions along soil deep layers (Senga et al., 120 2011; Xiang et al., 2015). The impacts of plant invasion on the distribution of total soil C 121 accumulated among the various soil fractions and the further relationships of these impacts 122 on C dynamics with other soil traits, such as C:N:P ratios, are also unclear (Sardans & 123 Peñuelas, 2012). We hypothesized that changes in soil C storage along the soil profile could 124 be related to different plant species dominating above-ground. The knowledge of the 125 variation in C accumulation along the vertical soil profile and its causes is crucial for 126 understanding the capacity of wetlands to act as carbon sinks in absolute amounts (Craft, 127 2007). 128 We aimed (i) to determine whether the success of plant invasion of the same invasive 129 species at different sites was based on a similar strategy of C, N, and P use, and (ii) to compare 130 the differences of soil traits and plant-soil system C, N, and P stocks between successful 131 recently established invasive communities of Spartina alterniflora in two very different 132 native communities: native tallgrass and mangrove communities with respect pure non- 133 invaded native communities in different areas along subtropical and tropical coastal wetlands 134 of China. 135 136 MATERIALS AND METHODS 137 Study area 138 We studied the coastal wetlands of central-south China, where today plant invasion effect is 139 one of the most important problems (Figure 1). In 1979, Spartina alternifolia was introduced 140 into China from the United States for the purpose of accumulating silt, thereby protecting 141 beaches and berms. However, S. alterniflora was identified as an efficient invasive plant, 6 142 decreasing the area of indigenous vegetation communities, transforming the habitat, and 143 jeopardizing the survival of the animal species in the intertidal zone, and ultimately 144 producing a series of ecological and economic hazards to the area (Yang et al., 2009). S. 145 alterniflora is the only coastal wetland invasive plant identified by the State Environmental 146 Protection Administration of China since 2003. S. alterniflora has an apparently superior 147 breeding ability and adaptability than several native wetland species. Thus, it has spread 148 wildly and invaded and replaced coastal native plant communities where it forms mono- 149 species communities. Today, S. alterniflora grows from the North of China from the Yalu 150 River estuary to the South in Guangxi Beibu Gulf. The distribution area accounts for about 151 2% of the wetland area in China (Lu & Zhang, 2013). China is the country with the largest 152 outbreak of S. alterniflora in the world, especially in its subtropical coastal wetland. The 153 distribution area of S. alterniflora reaches more than 92.44% of the total invaded areas in the 154 subtropical coastal wetland (Lu & Zhang, 2013). Therefore, the subtropical coastal wetland 155 is an ideal area for the study of S. alterniflora invasion. 156 157 Sample collection and measurements 158 Soil samples were collected during the plants’ main growing season (June to August) in 2015 159 from eight locations in China’s subtropical and tropical coastal wetlands (Figure 1). We 160 sampled soils in well stablished Spartina alterniflora communities in sites originally 161 dominated by Phragmite australis (grass), Kandelia obovata (mangrove), Avicennia marina 162 (mangrove), Spartina alterniflora (invaded from 7 to 15 years ago), and also collected soils 163 form native plants communities growing at these sites for more than 30 years. In our study, 164 for the S. alterniflora-invaded Phragmite australis site, the native plants were completely 165 replaced by S. alterniflora. However, for the S. alterniflora-invaded mangrove site, the 7 166 invasion has begun in mangrove community margins and empty spaces within the mangrove 167 community, and it is currently spreading its cover by all the mangrove community space. 168 Three plots were randomly established in each community type, and soil profiles (width, 1 169 m; length, 0.6 m; depth, 0.4 m) were excavated. The samples were collected with a small 170 sampler (length, 0.4 m; diameter, 0.1 m) from each of four soil layers (0–10, 10–20, 20–30, 171 and 30–40 cm) at the center and both ends of the soil pits. These three samples from each 172 layer were bulked to form one composite sample per layer. A total of 240 soil samples (ten 173 sampling site × two communities × three plots × four soil layers) were thus collected. The 174 core samples were divided into two parts, with one part unprocessed for the measurement of 175 soil MBC and DOC, and the other part air-dried and finely grounded in a ball mill after the 176 removal of all roots, and visible plant remains used for the determination of total SOC and 177 LOC. Total SOC was measured with a Vario EL III Elemental Analyzer (Elementar Scientific 178 Instruments, Hanau, Germany), DOC by extracting the soils with deionized water (1:5 ratio) 179 and measuring the C concentration using a TOC-V CPH total C analyzer (Shimadzu 180 Scientific Instruments, Kyoto, Japan), MBC by fumigation-extraction (Lu, 1999), and LOC 181 by digestion with 333 mM KMnO4 (Xu et al., 2011; Wang et al., 2015c). 182 Bulk density was measured for the three bulked cores (5 cm diameter, 3 cm depth) 183 collected from each soil layer. Soil salinity, pH, and particle-size distribution were measured 184 by a DDS-307 salinity meter (Boqu Scientific Instruments, Shanghai, China), an 868 pH 185 meter (Orion Scientific Instruments, Minnesota, USA), and a Master Sizer 2000 Laser 186 Particle Size Analyzer (Malvern Scientific Instruments, Suffolk, UK), respectively. Soil total 187 N concentration was determined with a Vario EL III Elemental Analyzer (Elementar 188 Scientific Instruments, Hanau, Germany), and total soil P concentration was determined by 189 perchloric-acid digestion followed by ammonium-molybdate colorimetry and measurement 8 190 using a UV-2450 spectrophotometer (Shimadzu Scientific Instruments, Kyoto, Japan). 191 Tallgrass above-ground biomass in each plot was collected from a randomly selected 192 quadrat (10 × 10 m), and the above-ground biomasses were collected from a selected center 193 sub-quadrat (1 × 1 m). For the mangrove, we only collected some above-ground materials, 194 not all of the above-ground biomass. At the same time, we determined the mangrove diameter 195 at the breast height and the height and density of all plants. Also, the above-ground mangrove 196 biomass of each plant population was calculated by a previously reported equation (Tam et 197 al., 1995). 198 All plant material was gently washed with deionized water and then oven-dried to a 199 constant mass (80 °C for 24–36 h) and weighed. The concentrations of C and N of the plants 200 were determined using a Vario EL III Elemental Analyzer (Elementar Scientific Instruments, 201 Hanau, Germany). Total plant P concentrations were all determined by the method of 202 colorimetrically using a UV-2450 spectrophotometer (Shimadzu Scientific Instruments, 203 Kyoto, Japan) at a wavelength of 700 nm (Lu, 1999). C, N, and P stocks in the plant were 204 calculated by the biomass multiplied by the nutrient concentration. 205 206 Statistical analyses 207 We performed general mixed models (GLM) with plant type (invasive versus native) as an 208 independent fixed categorical variable, site as an independent random factor, and soil traits 209 plant above-ground biomass, and C, N, and P stocks in plant above-ground biomass as 210 continuous dependent variables. We used the “nlme” (Pinheiro et al., 2016) R package with 211 the “lme” function. If the variable was not normally distributed, it was log-transformed. We 212 chose the best model for each dependent variable based on the Akaike information criterion. 9 213 We used the MuMIn (Barton, 2012) R package in the mixed models to estimate the 214 percentage of the variance explained by the mixed models. We conducted Tukey’s post hoc 215 tests to detect significant differences in the analyses for more than two communities using 216 the “multcomp” (Hothorn et al., 2013) R package with the “glht” function. 217 We also performed other multivariate statistical analyses. We determined the overall 218 differences in the changes of the soil C fractions, N, and P concentrations, stoichiometric 219 ratios, salinity, pH, and temperature in the species-specific plant communities using general 220 discriminant analysis (GDA), including the component of the variance due to the different 221 soil layers as an independent categorical variable. Discriminant analyses consist of a 222 supervised statistical algorithm that derives an optimal separation between groups established 223 a priori by maximizing between-group variance while minimizing within-group variance 224 (Raamsdonk et al., 2001). GDA is thus an adequate tool for identifying the variables most 225 responsible for the differences among groups, while controlling the component of the 226 variance due to other categorical variables. The GDAs were performed using Statistica 6.0 227 (StatSoft, Inc. Tulsa, USA). 228 229 RESULTS 230 Spartina alterniflora invasion of tallgrass wetlands 231 Compared to the original native communities of Phragmite australis, the Spartina 232 alterniflora communities that currently occupy wetland areas previously inhabited by the 233 native tallgrass P. australis had higher soil C in microbial biomass, total soil C and P content, 234 soil C:N ratio, and P content in above-ground biomass, but lower soil N:P ratio and C stocks 235 in above-ground plant biomass (Table 1, Figures 2 and 3). Our data analyses provide 236 evidences of higher soil C (approx. 18.8 versus 17.5 t C ha-1) in the invaded than native 10 237 community but smaller C stocks in aboveground biomass (approx. 6 versus 8 t C ha-1) in the 238 invaded than in native communities, thus resulting in a trend to less total C stored in the plant- 239 soil system, irrespective of the C stored in below-ground biomass. Additionally, our data 240 indicate higher amounts of P stored in soil (approx. 0.07 t P ha-1) and in aboveground biomass 241 (approx. 1.5 g m-2, that is 0.015 t P ha-1) in invasive than in native communities. Therefore, 242 without knowing the P stored in the belowground biomass, the data showed a total higher P 243 stored in the invasive than native communities. 244 The GDA analysis with soil data, comparing the communities dominated by 245 Phragmites australis with those dominated by Spartina alterniflora, showed no differences 246 in main soil features (Squared Mahanalobis = 2.46, F = 1.01, P = 0.46) between the 247 Phragmites australis and Spartina alterniflora communities. 248 249 Spartina alterniflora invasion of mangrove 250 In our study, the Spartina alterniflora communities that currently occupy areas in the margins 251 of mangrove community and in the empty spaces within the mangrove community, which 252 previously were occupied by native mangrove communities (mostly previously inhabited by 253 Kandelia obovata or Avicennia marina), had lower soil organic carbon, total soil C and P 254 content, and soil C:N, C:P, and N:P ratios (Table 2). Moreover, Spartina alterniflora 255 communities also had lower C, N, and P stocks in the above-ground plant biomass (only 256 compared with Avicennia-dominated mangroves) and lower aboveground biomass than 257 native mangrove communities (Figures 2 and 3). Hence, data analyses indicate higher stocks 258 of soil C (approx. 2.6 t C ha-1), N (approx. 0.17 t N ha-1), and P (approx. 0.05 t P ha-1) and 259 also higher stocks in the above-ground biomass of C (approx. 2.2 t C ha-1), N (approx. 0.65 t 260 N ha-1), and P (approx. 0.08 t P ha-1) in the native than invaded community (Figures 2 and 3). 11 261 Thus resulting in a trend to substantially less C, N, and P stored in the invaded plant-soil 262 system relative to the native-dominated mangrove communities, irrespective of the C, N, and 263 P stored in below-ground biomass. 264 In contrast, in this case, the GDA analysis of the soil data, comparing the Spartina- 265 dominated with the native mangrove communities, revealed significant differences (Squared 266 Mahalanobis = 9.84, F = 17.3, P < 0.0001) arising after the invasion (Figure 4). 267 268 DISCUSSION 269 General descriptive results 270 We have not observed important changes in C:N and C:P soil ratios along the soil profile. 271 This is an interesting and particular result of these studied wetland soils, which is probably 272 related to the sediment loadings that these soils continuously receive and the anoxic 273 conditions that prevent a fast and complete decomposition of the litter, which is incessantly 274 covered by new sediments and progressively buried to deeper soil layers. A trend to uniform 275 organic carbon contents along the vertical soil profile in wetlands has been previously 276 observed in soils receiving regular loads of sediments (Alongi et al., 2001; Senga et al., 2011), 277 which is related to the physical protection of soil organic carbon linked to sedimentation 278 typical from seasonally-flooded wetlands (Maynard et al., 2011). 279 280 The relationships between species invasion in tallgrass wetland communities and soil 281 traits and plant-soil C, N, and P stocks 282 Our results show that the invasion of Spartina alterniflora replacing Phragmites australis, 283 when comparing adult communities of both species, had no clear impacts on the overall C 284 storing capacity of the plant-soil systems. Although the invaded community had more C in 12 285 the soil, there was a correspondingly lower amount of C stored in the above-ground biomass. 286 Previous studies conducted elsewhere in Chinese wetlands and investigating the impact of S. 287 alterniflora on soil properties after replacing tallgrass communities dominated by other 288 native species, such as Cyperus malaccensis or Suaeda salsa, also showed a trend to 289 increasing soil C stocks with time since invasion (Chen et al., 2015; Jin et al., 2017; Liu et 290 al., 2017). In light of these previous results, our data appear consistent with the general view 291 that invasive processes in wetlands tend to maintain or even increase C stocks in the soil, as 292 observed at the global level (Liao et al., 2008). 293 Reports addressing the impacts of S. alterniflora on soil stocks of N and P and on the 294 stocks of C, N, and P in plant biomass are rare. After the replacement of P. autralis by S. 295 alterniflora, we found a general increase in P stocks both in the soil and above-ground 296 biomass, but not in N, thus resulting in a decrease in soil and plant N:P ratios. To our 297 knowledge, so far there are very few equivalent studies with which we can compare our 298 results. Likewise, Wang et al. (2015a) found a decrease in N:P ratios in a wetland area of the 299 Minjiang River, where S. alterniflora was invading the native Cyperus malaccensis tall grass 300 communities, although they did not observe any effects of invasion on soil P concentrations. 301 The higher P contents in soil observed after the invasion was mainly due to the higher 302 bulk density in the invaded than in the native communities. Bulk density has been positively 303 correlated with organic C, N, and P concentrations in similar wetland areas in the soils below 304 Carex sp. and Phragmites australis (Peng et al., 2005). Differences and fluctuations in bulk 305 density have also been frequently associated with the success of invasive plant species (Miller 306 et al., 2006; Lortie & Cushman, 2007; Pande et al., 2007). Summarizing, the successful 307 invasion of Spartina alterniflora replacing Phragmites australis in Chinese wetlands should 308 not produce great impacts on their C storing capacity, but it could increase soil and ecosystem 13 309 P-storing ability due to the higher plant P concentrations and also due to the increase in soil 310 bulk density, further decreasing the N:P ratios at an ecosystem level with likely consequences 311 for the whole soil and above-ground trophic chains. 312 313 The relationships between species invasion in mangrove communities and soil traits and 314 plant-soil C, N, and P stocks 315 The lower soil bulk densities in S. alterniflora-invaded communities relative to the native 316 mangrove communities were mainly related to a lower sand content in the latter. The success 317 of S. alterniflora was thus associated with decreases in bulk density and total soil C, N, and 318 P contents per unit of surface area. 319 As in other studies that have observed differences in bulk density and elemental 320 composition between soils in invaded and native communities (Wang et al., 2018), the causes 321 of these differences in our study were unclear. Nevertheless, this result suggests that the 322 scarcer root system of a tallgrass community, compared with the more consistent root system 323 of a mangrove stand, could have a lower ability to retain small-size particles in the soil, such 324 as clay. 325 In this case, the large differences in above-ground biomass between the native 326 mangrove stands compared to the Spartina alternifora-dominated tallgrass community can 327 easily explain the great losses in C, N, and P stocks in the soils after the invasion. Changes 328 in soil traits due to plant invasion, such as SOC concentration (Zhou et al., 2015), texture 329 (Uselman et al., 2014; Haubensak et al., 2014), and other soil parameters (Blank et al., 2015), 330 have been correlated with differences in plant or root biomass and/or shoot:root ratio between 331 native and invaded plant communities. In fact, soil properties such as texture can change in 332 wetlands within 10 years, as observed in other studies (Craft et al., 1999 and 2002). Other 14 333 works have also observed changes in soil bulk density (Zhang et al., 2009), C contents 334 (Koutika et al., 2007; Koteen et al., 2011; Throop et al., 2013; Yu et al., 2014), and several 335 other traits (Yu et al., 2014; Souza-Alonso et al., 2015) as a consequence of successful plant 336 invasion. The similar SOC values along the vertical soil profile shown earlier suggest that 337 these wetlands are sinks of sediments, reinforcing the idea that sites with different vegetation 338 cover with different structure and root system size can change soil structure by favoring 339 sedimentation of different particle size classes. Thus, our study is consistent with the 340 observation that soil texture is a sensitive factor during plant invasion, as has been observed 341 in several previous studies (Craft et al., 1999 and 2002; Surrette & Brewer, 2008; Zhang et 342 al., 2009). The evidence emerging from many previous investigations generally indicates that 343 the differences in soil properties between invaded and original native communities are 344 associated with the distinct structural and functional traits of the invasive versus the native 345 species; namely nutrient uptake, root system size and structure, litter production, foliar 346 nutrient resorption, and growth rate (Lindsay & French, 2005; Titus, 2009; Sardans & 347 Peñuelas, 2012; Novoa et al., 2014). 348 Nevertheless, the replacement of mangrove woody ecosystems by tallgrass 349 formations is clearly associated with an extensive change in the physical and chemical traits 350 of the whole ecosystem, with biomass loss, reductions in soil bulk density, and depletions of 351 C, N, and P stocks in the plant-soil system. These findings suggest that an eventual future 352 regeneration process would be difficult because the ecosystem has become a lot less fertile 353 and with sandy soil, conditions that are unfavorable for the reestablishment of the original 354 woody community, whose higher biomass requires greater resources that largely exceeds the 355 supply capacity of the current invaded system. Additionally, the lower ability to retain N and 356 P has negative consequences for water quality since it promotes eutrophication processes. 15 357 Moreover, these elements combined also worsen the overall C and nutrients storing capacity 358 of these wetlands and should be considered as a general process reducing the environmental 359 quality of the ecosystem. 360 361 FINAL REMARKS AND CONCLUSION 362 We show that the strong ecosystem impacts on soil characteristics, biomass accumulation, 363 and plant-soil system C, N, and P stocks after the successful invasion of S. alterniflora are 364 clearly different depending on the original native community being replaced. 365 When replacing the native grass P. australis, S. alterniflora did not produce a change 366 in soil traits, biomass accumulation or plant-soil C and N storing capacity. However, a higher 367 ability to store more P in both the soil and stand plant biomass suggests the possibility of a 368 decrease in ecosystem N:P ratios, which may imply future consequences for the whole below- 369 and above-ground trophic chains. Secondly, a higher efficiency to absorb P might improve 370 the P-filtering performance by the ecosystem, thus better preventing the eutrophication of 371 water. 372 However, the replacement of the native mangrove communities by S. alterniflora 373 resulted in several critical impacts and involved a serious environmental problem. These 374 included an enrichment of sand in soil with a consequent loss of nutrient retention capacity, 375 the loss of great amounts of C, N, and P out of the plant-soil system with a general loss of 376 plant production and decay of water quality, thereby exacerbating potential eutrophication 377 processes. Moreover, the loss of C and nutrient decreased the overall fertility of the system, 378 strongly hampering the future re-establishment of the original woody mangrove communities. 379 380 ACKNOWLEDGMENTS 16 381 The authors would like to thank Haifeng Fu, Dengzhou Gao, Xiao Lin, Youyang Chen, and 382 Xiaoxuan Chen for their assistance with field sampling and laboratory analysis. Funding was 383 provided by the National Science Foundation of China (41571287; 41601102), Ministry of 384 Science and Technology Basic Work Special Project of China (2013FY111805), Research 385 Project of Public Institute of Fujian Province (2018R1034-1), Outstanding Young Research 386 Talents in Higher Education of Fujian Province (2017), Spanish Government grant 387 CGL2016-79835, Catalan Government grant SGR 2017-1005, and European Research 388 Council Synergy grant ERC-SyG-2013-610028, IMBALANCE-P. 389 390 CONFLICS OF INTEREST 391 The authors declare no conflicts of interest. 392 393 RERERENCES 394 Alongi, D. M., Wattayakorn, G., Pfitzner, J., Tirendi, F., Zagorskis, I., Brunskill, G. J., 395 Davidson, A., & Clough, B. F. (2001). Organic carbon accumulation and metabolic 396 pathways in sediments of mangrove forests in southern Thailand. Marine Geology, 179, 397 85–103. 398 Andreetta, A., Huertas, A. D., Lotti, M., & Cerise, S. (2016). Land use changes affecting soil 399 organic carbon storage along a mangrove swamp rice chronosequence in the Cacheu 400 and Oio regions (northern Guinea-Bissau). Agriculture, Ecosystems and Environment, 401 216, 314–321. 402 Barton, K. (2012). MuM. In: Multi-model inference. R package version 1.7.2. http://CRAN. 403 R-283 project.org/package=MuMIn. at http://cran.r-project.org/ package=MuMIn>. 404 Bianchi, T. S., Allison, M. A., Zhao, J., Li, X. X., Comeaux, R. S., Feagin, R. A., & 17 405 Kulawardhana, R. W. (2013). Historical reconstruction of mangrove expansion in the 406 Gulf of Mexico: Linking climate change with carbon sequestration in coastal wetlands. 407 Estuarine Coastal & Shelf, 119, 7–16. 408 Blank, R. R., Morgan, T., & Allen, F. (2015). Suppression of annual Bromus tectorum by 409 perennial Agropyron cristatus: roles of soil nitrogen availability and biological soil 410 space. AoB Plants 7. In Press. DOI:10.1093/aobpla/plv006. 411 Bonan, G. B., & Van Cleve, K. (2011). Soil temperature, nitrogen mineralization and carbon 412 source-sink relationships in boreal forests. Canadian Journal of Forest Research, 22, 413 629–639. 414 Burden, A., Garbutt, R. A., Evans, C. D., Jones, D. L., & Cooper, D. M. (2013). Carbon 415 sequestration and biogeochemical cycling in a saltmarsh subject to coastal managed 416 realignment. Estuarine Coastal & Shelf Science, 120, 12–20. 417 Chen, H. L., Zhou, J. M., & Xiao, B. H. (2010). Characterization of dissolved organic matter 418 derived from rice straw at different stages of decay. Journal of Soil and Sediments, 10, 419 915–922. 420 Chen, Y. P., Chen, G. C., & Ye, Y. (2015) Coastal vegetation invasion increases greenhouse 421 gas emission from wetland soils but also increases soil carbon accumulation. Science 422 of the Total Environmental, 526, 19–28. 423 Craft, C., Reader, J., Sacco, J. N., & Broome, S. W. (1999) Twenty-five years of ecosystem 424 development of constructed Spartina alterniflora marshes. Ecological Applications, 425 9, 1405–1419. 426 Craft, C., Broome, S., & Campbell, C. (2002) Fifteen years of vegetation and soil 427 development after brackish-water marsh creation. Restoration Ecology, 10, 248–258. 18 428 Craft, C. (2007). Freshwater input structures soil properties, vertical accretion, and nutrient 429 accumulation of Georgia and U.S. tidal marshes. Limnlogy and Oceanography, 52, 430 1220–1230. 431 Daehler, C. C. (2003). Performance comparisons of co-occuring native and alien invasive 432 plants: implications for conservation and restoration. Annual Review of Ecology 433 Evolution and Systematics, 34, 183–211. 434 Funk, J. L., & Vitousek, P. M. (2007). Resource-use efficiency and plant invasion in low- 435 resource systems. Nature, 446, 1079–1081. 436 González, A. L., Kominoski, J. S., Danger, M., Ishida, S., Iwai, N., & Rubach, A. (2010). 437 Can ecological stoichiometry help explain patterns of biological invasion? Oikos, 438 119, 779–790. 439 Gu, L., Post, W. M., & King, A. W. (2004). Fast labile carbon turnover obscures sensitivity 440 of heterotrophic respiration from soil to temperature: a model analysis. Global 441 Biogeochemical Cycles, 18, GB1022. 442 Hanke, A., Cerli, C., Muhr, J., Borken, W., & Kalbitz, K. (2013). Redox control on carbon 443 mineralization and dissolved organic matter along a chronosequence of paddy soils. 444 European Journal of Soil Science, 64, 476–487. 445 Haubensak, K. A., D’Antonio, C. M., Embry, S., & Blank, R. (2014). A comparison of 446 Bromus tectorum growth and mycorrhizal colonization in salt deserts vs. sagebrush 447 habitats. Rangeland Ecology & Management, 67, 275–284. 448 Hothorn, T., Bretz, F., & Wesrfall, P. (2013). Package “mulcomp” (WWW document). U.R.L. 449 http://cran.stat.sfu.ca/web/packages/mulcomp/mulcomp.pdf. (accessed 19.12.17). 450 Howe, A. J., Rodriguez, J. F., & Saco, P. M. (2009). Surface evolution and carbon 451 sequestration in disturbed and undisturbed wetland soils of the Hunter Estuary, 19 452 southeast Australia. Estuarine, Coastal, and Shelf Science, 84, 75–83. 453 IPCC. (2000). Global Carbon Cycle Overview. Summary for Policymakers. Land Use, Land- 454 Use Change, and Forestry. IPCC Special Reports. Watson, R. T., Noble, I. R., Bolin, 455 B., Ravindranath, N. H., Verarde, D. J., & Dokken, D. J. (Eds.). International Panel on 456 Climate Change. pp. 375. 457 Jin, B. S., Lai, D. Y. F., Gao, D. Z., Tong, C., & Zeng, C. S. (2017). Changes in soil organic 458 carbon dynamics in a native C-4 plant-dominated tidal marsh following Spartina 459 alterniflora invasion. Pedosphere, 27, 856–867. 460 Joiner, D. W., Lafleur, P. M., McCaughey, J. H., & Bartlett, P. A. (1999). Interannual 461 variability in carbon dioxide exchanges at a boreal wetland in the BOREAS northern 462 study area. Journal of Geophysical Research- Atmospheres, 104, 27663–27672. 463 Kimura, M., Murase, J., & Lu, Y. (2004). Carbon cycling in rice field ecosystems in the 464 context of input, decomposition, and translocation of organic materials and the fates of 465 their end products (CO2 and CH4). Soil Biology and Biochemistry, 36, 1399–1416. 466 Knoblauch, C., Maarifat, A. A., Pfeiffer, E. M., & Faefele, S. M. (2011). Degradability of 467 black carbon and its impact on trace gas fluxes and carbon turnover in paddy soils. Soil 468 Biology and Biochemistry, 43, 1768–1778. 469 Koteen, L. E., Baldocchi, D. D., & Harte, J. (2011). Invasion of non-native grasses causes a 470 drop in soil carbon storage in California grasslands. Environmental Research Letters, 471 6, 044001. 472 Koutika, L. S., Vanderhoeven, S., Chapius-Lardy, L., Dassonville, N., & Meerts, P. (2007). 473 Assessment of changes in soil organic matter after invasion by exotic plant species. 474 Biology and Fertility of Soils, 44, 331–341. 20 475 Liao, C. Z., Peng, R. H., Luo, Y. Q., Zhou, X. H., Wu, X. W., Fang, C. M., Chen, J. K., & 476 Li, B. (2008). Altered ecosystem carbon and nitrogen cycles by plant invasion: a 477 meta-analysis. New Phytologist, 177, 706–714. 478 Lindsay, E. A., & French, K. (2005). Litterfall and nitrogen cycling following invasion by 479 Chrysanthemoides moniifera ssp rotundata in coastal Australia. Journal of Applied 480 Ecology, 42: 556–566. 481 Liu, J. Q., Zeng, C. S., Chen, N. (2006). Research of Minjiang River estuary wetland. 482 Beijing: Science Press. 483 Liu, J. E., Han, R. M., Su, H. R., Wu, Y. P., Zhang, L. M., Richardson, C. J., & Wang, G. X. 484 (2017) Effects of exotic Spartina alterniflora on vertical soil organic carbon 485 distribution and storage amount in coastal salt marshes in Jiangsu, China. Ecological 486 Engineering, 106, 132–139. 487 Lu, J., & Zhang, Y. (2013). Spatial distribution of an invasive plant Spartina alterniflora 488 and its potential as biofuels in China. Ecological Engineering, 52, 175–181. 489 Lortie, C. J., & Cushman, J. H. (2007). Effects of a directional abiotic gradient on plant 490 community dynamics and invasion in a coastal dune system. Journal of Ecology, 95, 491 468–481. 492 Lu, R. K. (1999). Analysis methods of soil science and agricultural chemistry. Beijing: 493 Agriculture Science and Technology Press. 494 Matzek, V. (2011). Superior performance and nutrient-use efficiency of invasive plants 495 over non-invasive congeners in a resource-limited environment. Biological Invasions, 496 13, 3005–3014. 21 497 Maynard, J. J., Dahlgren, R. A., & O’Geen, A. T. (2011). Soil carbon cycling and 498 sequestration in a seasonally satured wetland receiving agricultural runoff. 499 Biogeosciences, 8, 3391–3406. 500 Mendelsohn, R., Emanuel, K., Chonabayashi, S., & Bakkensen, L. (2012). The impact of 501 climate change on tropical cyclone damage. Nature Climate Change, 2, 205–209. 502 Miller, M. E., Belnap, J., Beatty, S. W., & Reynolds, R. L. (2006). Performance of Bromus 503 tectorum L. in relation to soil properties, water additions, and chemical amendments 504 in calcareous soils of southwestern Utah, USA. Plant and Soil, 288, 1–18. 505 Mitsch, W. J., Bernal, B., Nahlik, A. M., Mander, U., Zhang, L., Anderson, C. J., Jorgensen, 506 S. E., & Brix, H. (2013). Wetlands, carbon, and climate change. Landscape Ecology, 507 28, 583–597. 508 Novoa, A., Rodríguez, R., Richardson, D., & Gonzàlez, L. (2014). Soil quality: a key factor 509 in understanding plant invasión? The case of Capobrotus edulis (L.) N.E.Br. Biological 510 Invasions, 16, 429–443. 511 Pande, A., Williams, C. L., Lant, C. L., & Gibson, D. J. (2007). Using map algebra to 512 determine the mesoscale distribution of invasive plants: the case of Celastrus 513 orbicularis in Southern Illinois, USA. Biological Invasions, 9, 419–431. 514 Peng, P., Zhang, W., Tong, C., Qiu, S., & Zhang, W. (2005). Soil C, N and P contents and 515 their relationships with soil physical properties in wetlands of Dongting lake floodplain. 516 Chinese Journal of Applied Ecology, 16, 1872–1878. 517 Peñuelas, J., Sardans, J., Rivas-Ubach, A., & Janssens, I. A. (2012). The human-induced 518 imbalance between C, N, and P in Earth's life system. Global Change Biology, 189, 5– 519 8. 22 520 Peñuelas, J., Poulter, B., Sardans, J., Ciais, P., van der Velde, M., Bopp, L., Boucher, O., 521 Godderis ,Y., Llusià, J., Nardin, E., Vicca, S., Obersteiner, M., & Janssens, I. A. (2013). 522 Human-induced nitrogen-Phosphorus imbalances alter natural and managed 523 ecosystems across the globe. Nature Communications, 4, 2064. 524 Piecuch, C. G., & Ponte, R. M. (2014). Mechanisms of global-mean steric sea level change. 525 Journal of Climate, 27, 824–834. 526 Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & Core, T. R. (2016). nlme: Linear and 527 Nonlinear Mixed Effects Models. R package version 3.1-126, http://CRAN. 528 Raamsdonk, L. M., Teusink, B., Broadhurst, D., Zhang, N. S., Hayes, A., Walsh, M. C., 529 Berden, J. A., Brudle, K. M., Kell, D. K., Rowland, J. J., Westerhoff, H. V., van Dam, 530 K., & Oliver, S. G. (2001). A functional genomics strategy that uses metabolome data 531 to reveal the phenotype of silent mutations. Nature Biotechnology, 19, 45–50. 532 Ramsar. (2013). The Ramsar Manual. 6th edition. Ramsar Convention Secretariat. 533 Switzerland. pp. 110. 534 Sardans, J., & Peñuelas, J. (2012). The role of plants in the effects of Global Change on 535 nutrient availability and stoichiometry in the plant-soil system. Plant Physiology, 160, 536 1741–1761. 537 Sardans, J., Rivas-Ubach, A., & Peñuelas, J. (2012). The C:N:P stoichiometry of organisms 538 and ecosystems in a changing world: A review and perspectives. Perspectives in Plant 539 Ecology, Evolution and Systematics, 14, 33–47. 540 Sardans, J., Bartrons, M., Margalef, O., Gargallo-Garriga, A., Janssens, I. A., Ciais, P., 541 Obersteiner, M., Sigurdsson, B. D., Chen, H. Y., & Peñuelas, J. (2017). Plant invasion 542 is associated with higher plant-soil nutrient concentrations in nutrient-poor 543 environments. Global Change Biology, 23, 1282–1291. 23 544 Sardans, J., & Peñuelas, J. (2014). Hydraulic redistribution by plants and nutrient 545 stoichiometry: Shifts under global change. Ecohydrology, 7, 1–20. 546 Schewe, J., Levermann, A., & Meinshausen, M. (2011). Climate change under a scenario 547 near 1.5 degrees C of global warming: monsoon intensification, ocean warming, and 548 steric sea level rise. Earth System Dynamics, 2, 25–35. 549 Schipper, L. A., & Sparling, G. P. (2011). Accumulation of soil organic C and change in C:N 550 ratio after establishment of pastures on reverted scrubland in New Zealand. 551 Biogeochemistry, 104, 49–58. 552 Senga, Y., Hiroki, M., Nakamura, Y., Watarai, Y., Watanable, Y., & Nohara, S. (2011). 553 Vertical profiles of DIN, DOC, and microbial activities in the wetland soil of Kushiro 554 mire, northeastern Japan. Limnology, 12, 17–23. 555 Smith, P. (2004). Soils as carbon sinks: the global context. Soil Use and Management, 20, 556 212–218. 557 Souza-Alonso, P., Guisande-Collazo, A., & Gonzalez, L. (2015). Gradualism in Acacia 558 dealbata link invasion: impacts on soil chemistry and microbial community over a 559 chronological sequence. Soil Biology and Biochemistry, 80, 315–323. 560 Surrette, S. B., & Brewer, J. S. (2008). Inferring relationships between native plant diversity 561 and Lonicera japonica in upland forests in north Mississippi, USA. Applied Vegetation 562 Science, 11, 205–214. 563 Throop, H. L., Laitha, K., & Kramer, M. (2013). Density fractionation and 13C reveal changes 564 in soil carbon following woody encroachment in a desert ecosystem. Biogeochemistry, 565 112, 409–422. 566 Tian, H., Chen, G., Zhang, C., Melillo, J., & Hall, C. (2010). Pattern and variation of C: N: P 24 567 ratios in China’s soils: a synthesis of observational data. Biogeochemistry, 98, 139–151. 568 Titus, J. H. (2009). Nitrogen-fixers Alnus and Lupinus influence soil characteristics but not 569 colonization by later successional species in primary succession on Mount St. Helens. 570 Plant Ecology, 203, 289–301. 571 Uselman, S. M., Synder, K. A., Leger, E. A., & Duke, S. E. (2014). First-Year establishment, 572 biomass and seed production of early vs. late seral natives in two medusahead 573 (Taeniatherum caput-medusae) invaded soils. Invasive Plant Science and Management, 574 7, 291–302. 575 Wang, W. Q., Sardans, J., Zeng, C. S., Zhong, C., Li, Y., & Peñuelas, J. (2014). Responses 576 of soil nutrient concentrations and stoichiometry to different human land uses in a 577 subtropical tidal wetland. Geoderma, 232-234, 459–470. 578 Wang, W., Wang, C., Sardans, J., Zeng, C., Tong, C., & Peñuelas, J. (2015a). Plant invasive 579 success associated with higher N-use efficiency and stoichiometric shifts in the soil- 580 plant system in the Minjiang River tidal estuarine wetlands of China. Wetlands 581 Ecology and Management, 23, 865–880. 582 Wang, W., Wang, C., Sardans, J., Tong, C., Jia, R., Zeng, C., & Peñuelas, J. (2015b). Flood 583 regime affects soil stoichiometry and the distribution of the invasive plants in 584 subtropical estuarine wetlands in China. Catena, 128, 144–154. 585 Wang, W., Lai, D. Y. F., Wang, C., Pan, T., & Zeng, C. (2015c). Effects of rice straw 586 incorporation on active soil organic carbon pools in a subtropical paddy field. Soil and 587 Tillage Research, 152, 8–16. 588 Wang, W., Sardans, J., Wang, C., Asensio, D., Bartrons, M., & Peñuelas, J. (2018). Species- 589 specific impacts of invasive plant success on vertical profiles of soil carbon 590 accumulation and nutrient retention in the minjiang river tidal estuarine wetlands of 25 591 China. Soil Systems, 2(1), 5. 592 Wetland China. (2014). The report on the second national wetland resources survey (2009– 593 2013). http://www.shidi.org/zt/2014xwfbh/. 594 Wissing, L., Kölbl, A., Vogelsang, V., Fu, J., Cao, Z., & Kögel-Knabner, I. (2011). Organic 595 carbon accumulation in a 2000-year chronosequence of paddy soil evolution. Catena, 596 8, 376–385. 597 Xiang, J., Liu, D. Y., Ding, W. X., Yuan, J. J., & Lin, Y. X. (2015). Invasion 598 chronosequence of Spartina alterniflora on methane emission and organic carbon 599 sequestration in a coastal salt marsh. Atmospheric Environment, 112, 72–80. 600 Xu, M., Lou, Y., Sun, X., Wang, W., Baniyamuddin, M., & Zhao, K. (2011). Soil organic 601 carbon active fractions as early indicators for total carbon change under straw 602 incorporation. Biology and Fertility of Soils, 47, 745–752. 603 Yang, S. G., Li, J. H., Zheng, Z., & Meng, Z. (2009). Characterization of Spartina alterniflora 604 as feedstock for anaerobic digestion. Biomass and Bioenergy, 33, 597–602. 605 Yu, F. K., Huang, X. H., Duan, C. Q., He, S. Z., Zhang, G. S., Liu, C. E., Fu, D. G., & Shao, 606 H. B. (2014). Impacts of Ageratina adenophora invasion on soil physical-chemical 607 properties of Eucalyptus plantation and implications for constructing agro-forest 608 ecosystem. Ecological Engineering, 64, 130–135. 609 Zhang, C. B., Wang, J., Qian, B. Y., & Li, W. H. (2009). Effects of the invader Solidago 610 Canadensis on soil properties. Applied Soil Ecology, 43, 163–169. 611 Zhao, F., Yang, G., Han, X., Feng, Y., & Ren, G. (2014). Stratification of carbon fractions 612 and carbon management index in deep soil affected by the Grain-to-Green Program in 613 China. PloS one, 9, e99657. 26 614 Zhou, L. Y., Yin, S. L., An, S. Q., Yang, W., Deng, Q., Xie, D., Ji, H. T., Ouyang, Y., & 615 Cheng, X. L. (2015). Spartina alterniflora invasion alters carbon exchange and soil 616 organic carbon in eastern salt Marsh of China. Clean-Soil Air Water, 43, 569–576. 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 27 658 Figure Captions 659 660 FIGURE 1 The location of the study area and sampling sites in China. 661 662 FIGURE 2 C, N, and P stocks (mean ± S.E.) in above-ground biomass of invasive tallgrass 663 wetland communities of Spartina alterniflora (in black) replacing tall grass native 664 Phragmites australis community wetlands (in red) and also replacing mangrove (dominated 665 by the native species Avicennia marina) communities (in green). 666 667 FIGURE 3 Above-ground biomass (mean ± S.E.) of invasive tallgrass wetland communities 668 of Spartina alterniflora (in black) replacing tall grass native Phragmites australis community 669 wetlands (in red) and replacing mangrove (dominated by the native species Kandelia obovata) 670 communities (in blue) and also replacing mangrove (dominated by the native species 671 Avicennia marina) communities (in green). 672 673 FIGURE 4 Results of the general discriminant analysis (GDA) with the different studied soil 674 variables as independent continuous variables, different sites as categorical controlling 675 independent variable, and community types (Mangrove and Spartina alterniflora 676 communities) as dependent grouping categorical variable. 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 28 705 TABLE 1 Results of the mixed models of sites where Spartina alterniflora has replaced 706 native Phragmites australis tallgrass with different studied plant and soil variables as 707 dependent factors, species invasion as fixed factor, and site and soil depth as random values 708 in the case of soil variables. SOC = soil organic carbon, DOC = dissolved organic carbon, 709 LOC = labile organic carbon, MBC = microbial biomass organic carbon. Significant 710 responses (P < 0.05) to S. alterniflora invasion is highlighted in bold type Dependent Mixed model variables lme(Dependent variable ~ invasion, data = dades, random = list(~1|site, ~1|soildepth)) Fixed Factor (invasion) Model results Native Invasive F P-value R2m R2c SOC (mg g-1) 21.2±0.6 20.9±0.6 0.28 0.60 0.0022 0.48 LOC (mg g-1) 3.69±0.24 3.63±0.17 0.055 0.82 0.00051 0.34 DOC (mg g-1) 118±8 121±10 1.44 0.23 0.00093 0.95 MBC (mg g-1) 300±22 336±24 13.2 <0.0001 0.012 0.94 Soil Total C 17.5±0.6 18.9±0.7 4.36 0.041 0.024 0.61 content (t ha-1) Soil Total N 1.21±0.05 1.22±0.06 0.016 0.90 0.0001 0.57 content (t ha-1) Soil Total P 0.783±0.027 0.850±0.033 11.3 0.0014 0.027 0.83 content (t ha-1) Soil C:N 15.1±0.7 16.4±0.9 21.3 <0.0001 0.014 0.95 Soil C:P 22.4±0.5 22.4±0.5 0.0009 0.98 0.0001 0.17 Soil N:P 1.62±0.09 1.52±0.09 4.12 0.047 0.0072 0.88 Plant height 2.63±0.10 1.60±0.05 130 <0.0001 0.53 0.71 Plant density 90.9±9.2 155±9 220 <0.0001 0.20 0.93 (plant m-2) Soil [N] (g kg-1) 1.55±0.10 1.43±0.10 4.80 0.033 0.0080 0.88 Soil [P] (g kg-1) 0.944±0.013 0.931±0.012 1.99 0.16 0.0064 0.77 Soil pH 7.32±0.19 7.40±0.18 2.77 0.10 0.00089 0.98 Soil salinity (mS 1.92±0.29 1.97±0.30 0.20 0.65 0.00014 0.95 cm-1) Soil water content 41.9±1.5 40.8±1.6 2.29 0.14 0.0028 0.91 (%) Soil bulk density 0.851±0.037 0.912±0.039 7.09 0.01 0.014 0.86 (g cm-3) Soil Clay content 14.1±0.7 14.6±0.8 0.57 0.45 0.0023 0.71 (%) Soil silt content 65.1±1.3 62.4±1.4 14.8 <0.0001 0.022 0.90 (%) Soil sandy 20.7±1.8 23.0±2.1 5.37 0.024 0.0071 0.91 content (%) 711 712 29 713 TABLE 2 Results of the mixed models of sites where a Spartina alterniflora have replaced 714 a native mangrove community with different studied plant and soil variables as dependent 715 factors, species invasion as fixed factor and site and soil depth as random values in the case 716 of soil variables. SOC = soil organic carbon, DOC = dissolved organic carbon, LOC = labile 717 organic carbon, MBC = microbial biomass organic carbon. Significant responses (P<0.05) 718 to Spartina alterniflora invasion is highlighted in bold type 719 Dependent Mixed model variables lme(Dependent variable ~ invasion, data = dades, random = list(~1|site, ~1|soildepth)) Fixed Factor (invasion) Model results Native Invasive F P-value R2m R2c SOC (mg g-1) 16.6±0.5a 14.8±0.4b 41.9 <0.0001 0.034 0.86 LOC (mg g-1) 4.20±0.15 4.29±0.14 0.067 0.80 0.00033 0.17 DOC (mg g-1) 77.3±4.1 74.2±2.6 0.55 0.46 0.0025 0.25 MBC (mg g-1) 337±22 305±21 3.2 0.078 0.0060 0.69 Soil Total C 13.3±0.3 10.9±0.3 89.1 <0.0001 0.16 0.71 content (t ha-1) Soil Total N 1.20±0.03 1.03±0.02 45.8 <0.0001 0.10 0.63 content (t ha-1) Soil Total P 0.570±0.017 0.526±0.014 10.3 0.0017 0.020 0.68 content (t ha-1) Soil C:N 11.2±0.2 10.7±0.2 9.16 0.0029 0.016 0.71 Soil C:P 24.6±0.7 21.2±0.5 42.5 <0.0001 0.072 0.72 Soil N:P 2.30±0.10 2.06±0.07 15.2 <0.0001 0.017 0.81 Plant height 3.14±0.19 1.49±0.04 128 <0.0001 0.28 0.64 Plant density 1.29±0.07 100±10 202 <0.0001 0.35 0.71 (plant m-2) Soil [N] (g kg-1) 1.48±0.05 1.39±0.04 12.6 <0.0001 0.012 0.84 Soil [P] (g kg-1) 0.710±0.025 0.723±0.022 1.87 0.17 0.00086 0.92 Soil pH 6.14±0.13 6.65±0.13 24.7 <0.0001 0.035 0.76 Soil salinity (mS 4.21±0.22 4.29±0.25 0.11 0.74 0.00014 0.80 cm-1) Soil water 46.3±0.9 49.6±1.1 13.5 0.0003 0.026 0.68 content (%) Soil bulk 0.855±0.025 0.774±0.026 31.2 <0.0001 0.022 0.88 density (g cm-3) Soil Clay 13.8±0.6 14.6±0.5 1.81 0.18 0.0050 0.54 content (%) Soil silt content 58.5±2.1 60.5±2.0 5.26 0.023 0.0021 0.93 (%) Soil sandy 27.6±2.6 24.8±2.4 5.46 0.021 0.0028 0.91 content (%) 720 721 30