Both plant invasion and nitrogen (N) enrichment should have significant impact on mangrove ecosystems in coastal regions around the world. However, how N2O efflux in mangrove wetlands responds to these environmental changes has not been well studied. Here, we conducted a mesocosm experiment with native mangrove species Kandelia obovata, invasive salt marsh species Spartina alterniflora, and their mixture in a simulated tide rotation system with or without nitrogen addition. In the treatments without N addition, the N2O effluxes were relatively low and there were no significant variations among the three vegetation types. A pulse loading of exogenous ammonium nitrogen increased N2O effluxes from soils but the stimulatory effect gradually diminished over time, suggesting that frequent measurements are necessary to accurately understand the behavior of N-induced response of N2O emissions. With the N addition, the N2O effluxes from the invasive S. alterniflora were lower than that from native K. obovata mesocosms. This result may be attributed to higher growth of S. alterniflora consuming most of the available nitrogen in soils, and thus inhibiting N2O production. We concluded that N loading significantly increased N2O effluxes, while the invasion of S. alterniflora reduced N2O effluxes response to N loading in this simulated mangrove ecosystem. Thus, both plant invasion and excessive N loading can co-regulate soil N2O emissions from mangrove wetlands, which should be considered when projecting future N2O effluxes from this type of coastal wetland.
Citation: Jia D, Qi F, Xu X, Feng J, Wu H, Guo J, et al. (2016) Co-Regulations of Spartina alterniflora Invasion and Exogenous Nitrogen Loading on Soil N2O Efflux in Subtropical Mangrove Mesocosms. PLoS ONE 11(1): e0146199. doi:10.1371/journal.pone.0146199
Editor: Han Y.H. Chen, Lakehead University, CANADA
Received: August 24, 2015; Accepted: December 13, 2015; Published: January 4, 2016
Copyright: © 2016 Jia et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and supporting information.
Funding: This work was funded by National Science Foundation of China (2013CB956601) (http://www.nsfc.gov.cn/) to Guanghui Lin, by Ocean Public Fund Research Projects (201305021) (http://www.soa.gov.cn/) to Guanghui Lin, and by Shenzhen Key Laboratory for Coastal Ocean Dynamic and Environment (ZDSY20130402163735964) (http://hyxb.sz.tsinghua.edu.cn/) to Guanghui Lin. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Although nitrous oxide (N2O) contributes only 5% to estimated global warming potentials , its global warming strength is 265 times more powerful than CO2 over a 100-year time frame . N2O has also contributed to destruction of ozone in the stratosphere. During the past few decades, the global mean atmospheric concentration of N2O increased from 270 ppb in pre-industrial times to 324 ppb in 2011. The rapid increase in atmospheric N2O concentrations has gained much attention in quantifying N2O effluxes from various sources.
Coastal wetlands, such as mangroves in tropical and subtropical regions have been recognized as a major marine source of atmospheric N2O [4, 5]. In general, N2O effluxes from wetlands are related to several biological processes, mainly including nitrification (ammonium oxidation and nitrifier denitrification) , denitrification and nitrate-ammonification . These processes can be affected by many abiotic and biotic factors, such as soil temperature, oxygen level, and substrate availability (bioavailable carbon, ammonium and nitrate) in the soil [8, 9]. In addition, it is widely recognized that vegetation type can significantly affect N2O effluxes . Plants living in wetland environments can supply oxygen to the rhizosphere via aerenchymous tissue of plants, which creates oxidized microzones surrounding roots and rhizomes and favors nitrification and nitrification-denitrification reactions, facilitating N2O production [11–15]. Meanwhile, root exudates and debris can be used as an organic carbon source for microbes. The high activity of microbes could increase soil N mineralization  and promote N2O production. Therefore, the change in vegetation type, such as plant invasion, may have a potential effect on N2O effluxes.
Spartina alterniflora, a C4 salt marsh grass native to the east coast of the USA, was introduced to China in 1979. It has grown vigorously in China and spread over coastal wetlands since that time. During the last two decades, it has aggressively invaded habitats of native mangroves on the southern coast of China, especially in mangrove seedlings established areas [17–19], influencing the native species composition of mangrove ecosystems. As a result, it would alter soil available inorganic nitrogen (N) content and microbial activity [20, 21], which would further change N2O production and effluxes. So, it is essential to understand the change of N2O effluxes from mangrove wetlands following S. alterniflora invasion.
Globally, many mangrove ecosystems are also under significant threat from N pollution because of human activities, such as sewage discharge and surrounding aquaculture operation . Exogenous N loading has greatly changed functions of many mangrove wetlands in tropical and subtropical regions, from pristinely oligotrophic  to eutrophic . Usually, pristine mangrove soils, in N-limited environments, show low or negative N2O effluxes [25, 26]. Exogenous N loading often stimulate N2O effluxes by directly stimulating nitrification and denitrification in the soil or sediment [27, 28]. For example, the Futian mangrove swamp in China, an area with intense human activities, had the highest N2O effluxes compared to three other mangrove swamps with slightly anthropogenic disturbance . Similarly, the effluxes of N2O from the Puerto Rico mangrove swamp in the northeastern Caribbean Sea, where anthropogenic activities increase nitrogen loading to mangrove sediments, was much higher than those reported previously for intertidal estuarine sediments .
In the coastal wetlands, N loading from surrounding aquaculture ponds are always periodic. For example, during the common process of complete pond sediment cleaning, large quantities of sediment with high nutrient levels are washed out and discharged into the adjacent mangrove swamps in a very short period of time . N2O effluxes may vary greatly over time because of dynamic changes of soil inorganic N concentration after such exogenous N loading. In order to accurately estimate N loading effect on N2O effluxes, attention should be paid to dynamic changes of N2O effluxes after a pulse of exogenous N loading. However, to our knowledge, there has been limited research focusing on the dynamic change of N2O effluxes after an N-pulse loading in a wetland (but see ).
Although the previous research has pointed out that N2O effluxes in mangroves have positive response to excessive N loading, the response ability would be different between vegetation types . After N loading, the divergent growth rate between various species meant dissimilar uptake capabilities of ammonium nitrogen (NH4+-N) and nitrate nitrogen (NO3--N), which are the important substrates participating in nitrification and denitrification processes. The difference in the competition between microorganisms and plants for N may lead to a differential in soil N2O production [33, 34]. Thus, to gain an insight into the relationship between excessive N loading and N2O effluxes from mangrove wetlands after S. alterniflora invasion, the varying physiological responses of various plant species should also be considered.
In order to understand individual and possible interactive effects of S. alterniflora invasion and exogenous N loading on N2O effluxes from mangrove ecosystems, we conducted a mesocosm experiment with three vegetation types including the monocultures of K. obovata (a common native mangrove species in China) and S. alterniflora and their mixture and two N loading treatments (with or without N loading). The aims of the study were (1) to determine the dynamic change of N2O effluxes after a pulse N loading, (2) to test whether and how the invasion of S. alterniflora alters N2O effluxes, and (3) to investigate the interactive effect of plant invasion and N loading on mangrove N2O effluxes. We hypothesized that N loading will increase N2O effluxes from both mangrove and salt marsh mesocosms, but the invasive S. alterniflora mesocosms would have a lower increment of N2O effluxes after N loading due to larger N uptake compared with the native mangrove mesocoms.
Materials and Methods
The tide-system mesocosm experiment was conducted in a roof greenhouse at the Graduate School at Shenzhen, Tsinghua University, China (22°59′N, 113°97′E). The mesocosm system consisted of 18 cement tanks (2.40 m × 1.10 m × 0.50 m in volume) as experimental mesocosms and 2 cement tanks (7.50 m × 1.00 m × 0.70 m in volume) as seawater reservoirs (Fig 1). One reservoir was connected to all the mesocosms without N addition treatment and the other was connected to all the mesocosms with N addition treatment by pipelines of water inputting (used during tide flooding periods) and water outputting (used in tide falling periods). The artificial seawater at a salinity of 14 g l-1, representing an average salinity in a typical mangrove environment of southern China, was prepared by dissolving natural sea salts in tap water. Submersible pumps and timers were installed to simulate diurnal tides for all the mesocosms. For the transplanted plants to establish and acclimate, each mesocosm was conditioned under the tidal regime with one tidal cycle a day, 12 h high tide and 12 h low tide for 2 months before starting the experiment on August 31, 2012. After the formal start of the system operation, artificial seawater was pumped from the reservoir to each mesocosm to a depth of 5 cm between 14:00 and 20:00 (local time) to simulate high tide, then seawater was drained back by gravitational force to the reservoir during the low tide period. Tidal flux took about 30 minutes from the initiation of the ebb or flood tide. Tidal water in each reservoir was rotated for about 15 days as a water cycle, then drained and replenished by freshly prepared artificial seawater. In this way, we mimicked field conditions because daily tidal dynamic change might greatly change nitrification and denitrification potential, allowing us to determine whether the effects were robust with respect to natural perturbations.
Each mesocosm was filled with fresh mangrove soils. On June 5, 2012, the mangrove soils were collected from the mudflat connected to the Futian Mangrove Natural Reserve (22°51'N, 113°96'E), a typical mangrove swamp along the southern coast of Shenzhen. We collected the soils after receiving the permission of the Futian Mangrove Nature Reserve in this research. The substrate was sandy, consisting of 73.7% sand, 14.8% silt and 11.6% clay. After stones, benthic animals and plant residues in the soil were carefully removed, the soils was then used to fill each mesocosm to a depth of 30 cm (Fig 1). In the present study, K. obovata was chosen because it was the most dominant species among the eight true mangrove species in South China . The mature propagules of K. obovata and ramets of S. alterniflora were collected from the mangrove forest in Zhangjiangkou Mangrove Nature Reserve in Yunxiao, Fujian (23°56'N, 117°25'E). We collected the plants after receiving the permission of the Zhanjiang Mangrove National Nature Reserve in this research. The propagules of K. obovata were cultivated in the sand for 2 months before transplantation to the mesocosms. Then healthy K. obovata seedlings with approximately equal height of 15 cm (usually with 3–5 green leaves) were selected for this study. Meanwhile, the young ramets of S. alterniflora of approximately equal size with four true leaves were selected for this study. Each K. obovata seedling and ramet of S. alterniflora covered an area of 0.15 m × 0.15 m and 0.25 m × 0.25 m, respectively. This mimics typical seedling densities of these two species according to our observations in the field.
In this experiment, three vegetation types represented three stages of S. alterniflora invasion to K. obovata: (i) monoculture of K. obovata (no invasion), (ii) mixture of K. obovata and S. alterniflora (partial invasion), and (iii) monoculture of S. alterniflora (complete invasion) (Fig 1). All vegetation types were tested in the absence and presence of nitrogen addition. Ammonium chloride (NH4Cl) of 450 mg was added to synthetic seawater of 3000 ml in the reservoir connected to mesocosms with N addition treatment one day before the start of water cycle (i.e. one day before tide rising of the first day) in order to make NH4Cl dissolve thoroughly. During a typical 15 day water cycle period, no more NH4Cl was added into the reservoir. Because the N of water may be lost by transformation by microbes and volatilization in form of ammonia gas in the reservoir during low tide periods, we measured the actual rates of N loading, which were shown in Fig 2. The N levels we applied are within the range of sewage and wastewater and sludge from aquaculture ponds [36, 37].
Measurements of N2O effluxes
For the results reported in this paper, we selected a typical 15 day water cycle period. Artificial seawater pumping started at 14:00–20:00 on June 18, 2013 (Day 1) from a new pool of tidal water with desired salinities and N levels specified above. We started the collection of gas samples on the day after a new seawater change (i.e. on June 19, 2013) to investigate the effect of N loading on N2O effluxes during daytime of tide falling periods. During this water cycle period, the N2O effluxes in each mesocosm were quantified using a static chamber-gas chromatograph method on the day 2, 4, 6, 8, 10, 14 and 16. Each static chamber, which was made of PVC, covered a soil area of 0.025 m2 and had an internal volume of about 4000 ml. The open end of the chamber was inserted 3–5 cm into the soil with the air-tight valve open. Gas sampling was carried out 30 minutes after the chamber was inserted. Generally, five 8 ml gas samples of chamber air were collected by passing a hypodermic needle attached to a 10 ml glass syringe and was injected into pre-evacuated vials for laboratory analysis. Meanwhile, to minimize any effect of diurnal variation in N2O effluxes, samplings were conducted at roughly the same time of the day for the same mesocosm between 9:00 and 11:00. The deployment time was set to 45–90 min with sampling at 15–30 min intervals depending on the N2O efflux rates.
The N2O concentration of gas samples collected was analyzed within 24 h after sampling using a gas chromatograph (GC, Agilent 6820, Agilent Technologies, USA) equipped with an electron capture detector (mECD) and a HP- Plot/column (J&W GC Columns, Agilent Technologies, USA). The temperatures of the injector, column and detector were 50°C, 50°C and 300°C, respectively, with a carrier gas (N2) flow rate of 15 ml min-1. The relative standard deviation (RSD) of replicate standard measurements was 0.8%. During gas measurements, standard samples were analyzed with every 10 samples of determination to ensure that each sample run maintained a RSD below 3% in 12 h. N2O concentrations were quantified by comparing the peak areas of samples against the standard curves. N2O efflux rates were calculated from the linear change of the measured gas concentrations in the chamber headspace with known headspace volume and sampling time. Estimates of cumulative N2O emissions for each mesocosm were based on linear interpolation, with the sum of cumulative experiment period emissions representing one water cycle.
Sampling and analyses of water and soil
Everyday input water samples from each reservoir were collected before the start of tide flooding and output water samples from each mesocosm were collected before the start of tide falling during the experiment water cycle period. The water samples were transported to the laboratory and analyzed immediately. The samples were filtered through a 0.45 μm filter and analyzed for NH4+-N, NO3--N, nitrite nitrogen (NO2--N). We used standard procedures and methods for the determination, including the Sodium salicylate-sodium hypochlorite Method for NH4+-N, the UV Spectrophotometer Method for NO3--N and the α-Naphthylamine Method for NO2--N. Total inorganic nitrogen (TIN-N) was calculated by summing NH4+-N, NO3--N and NO2--N . Rates of N loading were calculated by multiplying N concentration of input water by the area and the flooding depth (5cm) of each mesocosm.
Three soil cores (0–15 cm) were collected around the chamber using hand-held PVC corers from each mesocosm at day 4 after gas sampling. Soil total carbon and nitrogen contents were measured by the Elementar (Vario ELⅢ, German). The NH4+-N and NO3--N contents in the KCl (2 M) extracts were also determined by the methods above. Both soil nitrification potential activity (PNA) and denitrification potential activity (PDA) were measured according to the method of Chen et al. (2012) .
Measurements of plant biomass
K. obovata biomass, measured on July 5, 2013, was evaluated by a non-destructive allometric technique . The relationships between biomass (dry weight) of each plant part, namely leaf, stem and plumular axis, and their leaf number (LN), and stem height (SH), propagule height (PH) and basal diameter (D) were obtained by harvesting 12 individuals of K. obovata from each mesocosm randomly. The best-fit equations for estimating biomass of leaves (LB), stems (SB) and propagules (PB) were LB = 0.32 LN − 0.89 (r2 = 0.89), SB = 0.23 × SH × D × D + 0.40 (r2 = 0.80), and PB = 0.52 × PH × D × D + 2.53 (r2 = 0.85), respectively. The total aboveground biomass was calculated by summing LB, SB and PB. The best-fit equations for estimating the total belowground biomass was BGB = 0.31 × AGB +1.16 (r2 = 0.85).
S. alterniflora biomass was measured also on July 5, 2013 but was evaluated by a non-destructive height–weight method . The relationships between aboveground biomass (B) of plant and their height (H) were obtained by harvesting 200 individuals of S. alterniflora from each mesocosm randomly. The best-fit equations for estimating aboveground biomass of was ln B = 1.44 ln H − 5.46 (r2 = 0.85).The best-fit equations for estimating BGB of were ln BGB = 0.61 × AGB + 34.88 (r2 = 0.86).
Plant samples were analyzed for N concentration with a Elementar elemental analyzer (Vario ELⅢ, German) and the total N accumulation was calculated by multiplying the N concentration by dry weight.
The main and interactive effects of sampling time, vegetation type and nitrogen addition treatment on N2O effluxes were tested with a parametric three-way analysis of variance (ANOVA). A one-way analysis of variance (ANOVA) performed to examine the significant differences of N2O effluxes between vegetation types for each period. One-way ANOVA was also used to analyze the differences in parameters of soil properties and plant traits between the treatments, followed by the least significant difference (Duncan) test at P < 0.05. Linear regression analyses were performed to evaluate the relationships of NH4+-N, NO3--N, NO2--N and TIN-N loading rates and N2O effluxes. The student's t-test was used to examine the significant difference of slopes and intercepts of linear regression lines of TIN-N and N2O effluxes, and NH4+-N and N2O effluxes, NO3--N and N2O effluxes and NO2--N and N2O effluxes. All statistical analyses were performed using SPSS 19.0 for Windows (SPSS Inc. USA).
Rate of N loading
In the treatment without N addition, NO3--N was the dominant dissolved N form of the input water and contributed 70.08% of TIN-N load. The rate of NO3--N loading declined gradually with time. With N addition, after the refreshment of tidal water, the rate of NH4+-N loading in the mesocosms decreased with time from 1.54 to 0.002 g m-2, However, the rate of NO3--N loading in the mesocosms increased to a peak by the day 8 (0.21 mg m-2), and then decreased gradually to 0.04 mg m-2. A similar variation pattern was observed for nitrite (NO2--N), with the highest rate (0.06 mg m-2) also on day 8. The rate of TIN-N loading generally decreased with time (Fig 2).
Soil properties and plant traits
In treatments without N addition, concentrations of soil total carbon, nitrogen, NH4+-N and NO3--N were not significantly affected by plant community composition, except that the NH4+-N concentration in mixture mesocosms was significantly lower than that in S. alterniflora mesocosms and K. obovata mesocosms. In treatments with N addition, NH4+-N concentrations differed among the three plant communities, ranging from 2.25 mg kg-1 in the S. alterniflora mesocosms, and 3.02 mg kg-1 in the mixture mesocosms to 3.27 mg kg-1 in the K. obovata mesocosms. PNA also varied with plant species. N loading increased the soil total carbon, NH4+-N and PNA significantly. However, soil total N, NO3--N and PDA were not significantly different between the N addition and without N addition mesocosms (Table 1).
Without N addition, no significant differences in aboveground biomass were found among vegetation types, while the belowground biomass in S. alterniflora mesocosms was 1.02 kg m-2 significantly higher than that in K. obovata mesocosms. N addition increased total biomass in the S. alterniflora, K. obovata and their mixture mesocosms by 55.26%, 176.76% and 148.78%, respectively. Under N addition condition, above- and below-ground biomass of S. alterniflora were evidently higher than that of K. obovata mesocosms. However, the ratio of aboveground to belowground biomass was 3.21 in K. obovata mesocosms, significantly higher than 0.99 in S. alterniflora mesocosms. Therefore, S. alterniflora was more productive compared to K. obovata (Table 1).
Without N addition, the N2O effluxes from K. obovata, the mixture and the S. alterniflora mesocosms ranged from 0.27 to 0.93, from 0.52 to 1.11 and from 0.32 to 0.70 μmol m-2 h-1 in the, respectively (Fig 3). N addition significantly increased mean N2O effluxes (Table 2). With N addition, the N2O effluxes ranged from 0.56 to 7.17, from 0.23 to 4.14 and from 0.19 to 4.10 μmol m-2 h-1 in the K. obovata, the mixture and the S. alterniflora mesocosms, respectively (Fig 3). N2O effluxes for all three vegetation types generally decreased with sampling time. In addition, there was a significant interaction of plant species × N fertilization on N2O effluxes (Table 2).
We further analyzed N2O effluxes over the 15-day period by dividing the sampling time into three periods according to the N loading rate and main species of N in the input water with N enrichment as shown in Fig 3. NH4+-N was the major species of N in the first period during day 1 to 6 with high N loading (TIN-N loading rates ranging from 0.51 to 1.87 g m-2); NO3--N was the major species of N in the second period during day 7 to day 13 with modest N loading (TIN-N loading rates ranging from 0.12 to 0.39 g m-2) and in the last period total N loading rates have decreased to background level during day 14 to day 16 (TIN-N loading rates ranging from 0.047 to 0.057 g m-2) (Fig 2).Without N addition, the N2O effluxes from K. obovata, the mixture and the S. alterniflora mesocosms did not vary significantly between vegetation types during all three periods, while there were trends that N2O effluxes were higher from mixture mesocosms than that from K. obovata and S. alterniflora mesocosms during all three periods. With N addition, compared to that from K. obovata mesocosms, the mean N2O effluxes were significantly lower from the S. alterniflora mesocosms in period 1 and 2 (Fig 4). However, in period 3, there were no significant difference among the vegetation types.
Relationship between N2O effluxes and N loading
When integrating all data either with or without N addition for analysis in each vegetation type respectively, N2O effluxes were found to significantly correlate to the rate of TIN-N, NH4+-N, NO3--N and NO2--N loading (P < 0.05), except for the relationship between N2O effluxes and NO3--N & NO2--N loading rate in the S. alterniflora mesocosms. The slopes of the linear relationship between rate of TIN-N loading and N2O effluxes for S. alterniflora and mixture mesocosms were significantly lower than that for K. obovata mesocosms by student's t-test (P < 0.05) (Fig 5). Thus, the increase rate of N2O effluxes by N addition was much lower in the S. alterniflora and mixture than native K. obovata mesocosms.
Our results from this mesocosm experimental study showed that a pulse loading of exogenous ammonium nitrogen resulted in a stimulatory effect on N2O effluxes but the effect gradually diminished over time and that the invasion of S. alterniflora could mitigate N2O effluxes following N loading. In addition, there was a significant antagonism effect between plant invasion and exogenous N loading on mangrove N2O effluxes. Our findings offer new insights into how plant invasion and N loading modulate N2O effluxes from mangrove wetlands.
Pulse effect of N addition on N2O effluxes
Compared to the treatment with N addition, we found that the N2O effluxes were relatively low (<1.50 μmol m-2 h-1) in the treatment without N addition in the K. obovata, S. alterniflora and their mixture mesocosms, which was consistent with the conclusions from the previous studies that natural coastal wetlands were weak sources for atmospheric N2O, such as undisturbed mangrove and salt marshes ecosystems [26, 27]. However, we did not observe any N2O sink in either pure S. alterniflora stand or the mixture of S. alterniflora and native mangrove seedlings, which was different from the results of Yuan, et al. (2014) . A likely explanation for this difference is that in our study soil nitrate levels never reached the very low level (< 1 mg N kg-1) .
N2O effluxes were significantly increased by the N loading in both mangrove and Spartina mesocosms, which was also found in salt marshes . Although N2O emissions accounted for a very low proportion of total N loading, ranging from 0.23–0.47% for the mesocosms with N addition treatment and from 2.22 to 3.55% for those without N loading (S1 Table), we could not neglect N loading effect on N2O emission, considering 265 times higher potential warming strength than CO2. Previous research also found relative lower proportions (ranging from 0.05–0.48% at N loading rate of 1.4 g m-2 in field experiment of salt marsh coastal wetland) . The stimulating effect of N loading on N2O effluxes results might be attributed to the fact that enzyme activities and the microbial population sizes including nitrifiers and denitrifiers in surface sediment had a dramatic increase after N addition [43, 44] and N loading supply the source of substrate for nitrification and denitrification in the soil or sediment. However, for the exgenous N loading, N2O effluxes may be underestimated or overestimated if they were not measured continuously after the N loading, as shown in this study that the stimulatory effect of N loading on N2O effluxes decreased gradually over time. A contributing factor to this result is that N2O fluxes were significantly related to N loading rates and these N loading rates usually decreased gradually after a pulse of nitrogen loading due partly to assimilation by wetland plants, transformation by microbes, and adsorption to sediments. So, N addition significantly increased the mean N2O effluxes in the first (Day 0–5) and second period (Day 6–10) in S. alterniflora mesocosms, K. obovata mesocosms and their mixture, but had no effect in the third period (Day 11–15). This result suggests that discretely or infrequently measuring is inadequate to accurately predict the behavior of exotic N-induced response of N2O effluxes.
Meanwhile, with N addition, we found the mean N2O effluxes in the first period were higher than in the second period. We infered that, except for the higher N loading rates in the first period, the higher N2O effluxes may also be attributed to varying reponses of N2O effluxes to different N species. Previous study pointed out that N2O effluxes were larger when NH4+ -N was the dominant substrate and nitrification was the main source of nitrous oxide compared to the condition when nitrate and nitrite predominate . In our study, major nitrogen species of input water were transformed from ammonium in the first period to nitrate in the second period with time, which could partly explain the reason for the different rates of N2O effluxes between the two periods.
Therefore, in order to get more accurate results especially in coastal wetlands which are frequently affected by tide, we should pay more attention to sampling time, daily N loading rates and N species of input water after N pulse addition. In addition, we added ammonium into seawater as the initial nitrogen species to assess the anthropogenic effects on N2O effluxes from coastal wetland, as it constitutes a major form of human inputs such as sewage and wastewater and sludge from aquaculture ponds. Future studies should also investigate the effects of other N species such as NO3--N on N2O effluxes.
Effects of plant invasion on N2O effluxes
Without N addition, there were no significant difference of N2O effluxes between the invasive S. alterniflora and native K. obovata mesocosms and more productive S. alterniflora did not result in higher N2O effluxes. This finding implied that the invasion of S. alterniflora played a limited role in mediating N2O effluxes from mangrove in N-limited condition. Previous research work have shown that fast-growing species such as S. alterniflora, a salt marsh species, increased N mineralization of soil to a greater extent than more conservative species such as K. obovata, a mangrove forest species [16, 45]. However, it has also been found that the rate of net N uptake was four- to six-fold higher for faster growing species than slower growing plant species (e.g., Poorter et al. (1991) ). Considering the high growth and high root uptake capacity of S. alterniflora , the mineralized N in soil could be largely consumed by S. alterniflora, which would result in no buildup of the soil mineral N pool. Therefore, without N addition, no difference in N2O effluxes between S. alterniflora and K. obovata mesocosms could be attributed to the similar concentration of soil N pool.
Meanwhile, we found significant antagonism effects of plant invasion and nitrogen loading on soil N2O effluxes. With N addition, the mean N2O effluxes from the invasive S. alterniflora were lower than that from native K. obovata mesocosms. In addition the increase rates of N2O effluxes by N addition were also lower in the S. alterniflora than native K. obovata mesocosms, indicating the invasion of S. alterniflora reduced N2O effluxes response to N loading in this simulated mangrove ecosystem. This result may be attributed to a crucial effect of vegetation type on the quality of N2O effluxes. On one hand, as important substrates participating in the processes of nitrification and denitrification by microbes , NH4+-N and NO3--N would be absorbed by vegetation, which may inhibit N2O production due to the competition between the plant and microorganism for soil inorganic N. In our study, compared to mangrove mesocosms, higher biomass and higher uptake of N in S. alterniflora mesocosms were found, which might mean higher photosynthetic capacity, leading to stronger competition ability for N source and lower soil inorganic N content for N2O production by microbes. On the other hand, in wetland, plants with aerenchyma can supply oxygen to the rhizosphere [12, 13], which can form oxidized zone in the rhizosphere that stimulates nitrification and nitrification-denitrification processes [15, 49], therefore facilitating N2O effluxes. Previous researches have shown that stronger oxidation activity in the rhizosphere of mangrove species like K. obovata than salt marsh species such as S. alterniflora [50, 51]. Indeed, we found significant higher PNA in K. obovata mesocosms compared to S. alterniflora mesocosms. Therefore, we inferred that the higher N2O effluxes in the K. obovata mesocosms than in S. alterniflora mesocosms could be partly attributed to the stronger oxidation in the rhizosphere of mangrove species.
The mixture mesocosms of S. alterniflora and K. obovata, reflecting partly invasion of S. alterniflora, had higher plant richness compared to the monoculture K. obovata or S. alterniflora. With N addition, both partly and complete invasion (in monoculture S. alterniflora and mixture mesocosms) reduced N2O effluxes from mangrove soil in the first period. However, only complete invasion of S. alterniflora reduced N2O effluxes from mangrove soil in the second period. Moreover, both complete and partly invasion did not reduce N2O effluxes from mangrove soil with N addition in the third period and without N addition for all three periods. In conclusion, S. alterniflora invasion could reduce N2O effluxes regardless of invasion degree under relatively higher nitrogen availability, but no relief effect of S. alterniflora invasion on N2O effluxes occurred under N-limiting condition. The results indicated that N addition can improve the effect of partly S. alterniflora invasion on N2O effluxes in mangrove, perhaps because N greatly increased S. alterniflora growth and competition with mangrove seedlings , leading to exhibition of S. alterniflora characteristic in mixture mesocosms with N addition.
Although substrate condition and results of our controlled mesocosm experiment were similar to field experiment , cautions must be exercised when extrapolating results from controlled mesocosm studies to field-scale processes.
N2O effluxes were significantly increased by the N loading in the K. obovata, the S. alterniflora and their mixture mesocosms, but the stimulatory effect of N loading on the N2O effluxes decreased gradually with time and dispeared when the N loading rates returned to the background level. The lower N2O effluxes and a weaker response to N addition in the S. alterniflora mesocosms than the K. obovata mesocosms were partly due to significantly higher growth of S. alterniflora, which could have led to lower increment of available N in the sediments for N2O production; and also due to higher oxidation capacity in the rhizosphere of K. obovata, which stimulated nitrification and nitrification-denitrification processes. Thus, S. alterniflora invasion into mangrove habitats could reduce N2O effluxes in case of amount of exogenous N loading, such as the input of sewage and wastewater and sludge from aquaculture ponds. These findings should be considered when projecting future N2O effluxes from global mangrove wetlands under the influences of both biological invasion and increasing exogenous N loading.
S1 Table. Mass balance of nitrogen (g m-2) in the mangrove and Spartina mesocosms under different N loading during the experiment period from June 5, 2013 to July 5, 2013.
We thank Zhonglei Wang, Cunxin Ning, Hui Chen, Qian Huang, Fang Liu and Jian Zhou for their assistance with the greenhouse experiments and gas sampling. We are also grateful to Weimin Song, Rashid Rafique, Junyi Liang, Zheng Shi and Jianyang Xia for editing the manuscript.
Conceived and designed the experiments: DJ YL GL. Performed the experiments: DJ FQ JF HW JG WL RP. Analyzed the data: DJ XX YL GL. Contributed reagents/materials/analysis tools: DJ FQ. Wrote the paper: DJ XX XZ YL GL.
- 1. Mosier A, Kroeze C, Nevison C, Oenema O, Seitzinger S, van Cleemput O. Closing the global N2O budget: nitrous oxide emissions through the agricultural nitrogen cycle—OECD/IPCC/IEA phase II development of IPCC guidelines for national greenhouse gas inventory methodology. Nutr Cycl Agroecosys. 1998;52(2–3): 225–248.
- 2. IPCC. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA.: Cambridge Univerisity Press; 2013.
- 3. Crutzen PJ, Ehhalt DH. Effects of nitrogen fertilizers and combustion on the stratospheric ozone layer. Ambio. 1977;6(2–3): 112–117.
- 4. Bange HW, Rapsomanikis S, Andreae MO. Nitrous oxide in coastal waters. Global Biogeochem Cycles. 1996;10(1): 197–207.
- 5. Nevison CD, Weiss RF, Erickson DJ. Global Oceanic Emissions of Nitrous-Oxide. J Geophys Res-Oceans. 1995;100(C8): 15809–15820.
- 6. Kowalchuk GA, Stephen JR. Ammonia-oxidizing bacteria: A model for molecular microbial ecology. Annu Rev Microbiol. 2001;55: 485–529. pmid:11544365
- 7. Knowles R. Denitrification. Microbiol Rev. 1982;46(1): 43–70. pmid:7045624
- 8. Wrage N, Velthof GL, Van Beusichem ML, Oenema O. Role of nitrifier denitrification in the production of nitrous oxide. Soil Biol Biochem. 2001;33(12): 1723–1732.
- 9. Weier KL, Doran JW, Power JF, Walters DT. Denitrification and the dinitrogen nitrous-oxide ratio as affected by soil-water, available carbon, and nitrate. Soil Sci Soc Am J. 1993;57(1): 66–72.
- 10. Sun H, Zhang C, Song C, Chang SX, Gu B, Chen Z, et al. The effects of plant diversity on nitrous oxide emissions in hydroponic microcosms. Atmos Environ. 2013;77: 544–547.
- 11. Edwards KR, Čižková H, Zemanová K, Šantrůčková H. Plant growth and microbial processes in a constructed wetland planted with Phalaris arundinacea. Ecol Eng. 2006;27(2): 153–165.
- 12. Rolletschek H, Bumiller A, Henze R, KOHL JG. Implications of missing efflux sites on convective ventilation and amino acid metabolism in Phragmites australis. New Phytol. 1998;140(2): 211–217.
- 13. Sasikala S, Tanaka N, Wah Wah H, Jinadasa K. Effects of water level fluctuation on radial oxygen loss, root porosity, and nitrogen removal in subsurface vertical flow wetland mesocosms. Ecol Eng. 2009;35(3): 410–417.
- 14. Inamori R, Gui P, Dass P, Matsumura M, Xu KQ, Kondo T, et al. Investigating CH4 and N2O emissions from eco-engineering wastewater treatment processes using constructed wetland microcosms. Process Biochem. 2007;42(3): 363–373.
- 15. Inamori R, Wang Y, Yamamoto T, Zhang J, Kong H, Xu K, et al. Seasonal effect on N2O formation in nitrification in constructed wetlands. Chemosphere. 2008;73(7): 1071–1077. doi: 10.1016/j.chemosphere.2008.07.064. pmid:18782640
- 16. Van der Krift TAJ, Berendse F. The effect of plant species on soil nitrogen mineralization. J Ecol. 2001;89(4): 555–561.
- 17. An SQ, Gu BH, Zhou CF, Wang ZS, Deng ZF, Zhi YB, et al. Spartina invasion in China: implications for invasive species management and future research. Weed Res. 2007;47(3): 183–191.
- 18. Chen L, Wang W, Zhang Y, Lin G. Recent progresses in mangrove conservation, restoration and research in China. J Plant Ecol. 2009;2(2): 45–54.
- 19. Zhang Y, Huang G, Wang W, Chen L, Lin G. Interactions between mangroves and exotic Spartina in an anthropogenically disturbed estuary in southern China. Ecology. 2012;93(3): 588–597. pmid:22624213
- 20. Hall SJ, Asner GP. Biological invasion alters regional nitrogen-oxide emissions from tropical rainforests. Glob Change Biol. 2007;13(10): 2143–2160.
- 21. Nagata O, Takakai F, Hatano R. Effect of Sasa invasion on global warming potential in Sphagnum dominated poor fen in Bibai. Phyton-Ann Rei Bot A. 2005;45(4): 299–307.
- 22. Vaiphasa C, De Boer WF, Skidmore AK, Panitchart S, Vaiphasa T, Bamrongrugsa N, et al. Impact of solid shrimp pond waste materials on mangrove growth and mortality: a case study from Pak Phanang, Thailand. Hydrobiologia. 2007;591(1): 47–57.
- 23. Vitousek PM, Aber JD, Howarth RW, Likens GE, Matson PA, Schindler DW, et al. Human alteration of the global nitrogen cycle: sources and consequences. Ecol Appl. 1997;7(3): 737–750.
- 24. Howarth RW, Sharpley A, Walker D. Sources of nutrient pollution to coastal waters in the United States: Implications for achieving coastal water quality goals. Estuaries. 2002;25(4): 656–676.
- 25. Allen DE, Dalal RC, Rennenberg H, Meyer RL, Reeves S, Schmidt S. Spatial and temporal variation of nitrous oxide and methane flux between subtropical mangrove sediments and the atmosphere. Soil Biol Biochem. 2007;39(2): 622–631.
- 26. Yuan J, Ding W, Liu D, Kang H, Freeman C, Xiang J, et al. Exotic Spartina alterniflora invasion alters ecosystem–atmosphere exchange of CH4 and N2O and carbon sequestration in a coastal salt marsh in China. Glob Change Biol. 2014;2015(21): 1567–1580.
- 27. Chen GC, Tam NFY, Ye Y. Summer fluxes of atmospheric greenhouse gases N2O, CH4 and CO2 from mangrove soil in South China. Sci Tot Environ. 2010;408(13): 2761–2767.
- 28. Kreuzwieser J, Buchholz J, Rennenberg H. Emission of methane and nitrous oxide by Australian mangrove ecosystems. Plant Biol. 2003;5(4): 423–431.
- 29. Corredor JE, Morell JM, Bauza J. Atmospheric nitrous oxide fluxes from mangrove sediments. Mar Pollut Bull. 1999;38(6): 473–478.
- 30. Wu H, Peng RH, Yang Y, He L, Wang WQ, Zheng TL, et al. Mariculture pond influence on mangrove areas in south China: Significantly larger nitrogen and phosphorus loadings from sediment wash-out than from tidal water exchange. Aquaculture. 2014;426: 204–212.
- 31. Moseman-Valtierra S, Gonzalez R, Kroeger KD, Tang J, Chao WC, Crusius J, et al. Short-term nitrogen additions can shift a coastal wetland from a sink to a source of N2O. Atmos Environ. 2011;45(26): 4390–4397.
- 32. Abalos D, Deyn GB, Kuyper TW, Groenigen JW. Plant species identity surpasses species richness as a key driver of N2O emissions from grassland. Glob Change Biol. 2014;20(1): 265–275.
- 33. Cheng X, Peng R, Chen J, Luo Y, Zhang Q, An S, et al. CH4 and N2O emissions from Spartina alterniflora and Phragmites australis in experimental mesocosms. Chemosphere. 2007;68(3): 420–427. pmid:17316757
- 34. Zhang YH, Wang L, Xie XJ, Huang LD, Wu YH. Effects of invasion of Spartina alterniflora and exogenous N deposition on N2O emissions in a coastal salt marsh. Ecol Eng. 2013;58: 77–83.
- 35. Tam N, Wong YS. Spatial variation of heavy metals in surface sediments of Hong Kong mangrove swamps. Environ Pollut. 2000;110(2): 195–205. pmid:15092834
- 36. Fontenot Q, Bonvillain C, Kilgen M, Boopathy R. Effects of temperature, salinity, and carbon: nitrogen ratio on sequencing batch reactor treating shrimp aquaculture wastewater. Bioresour Technol. 2007;98(9): 1700–1703. pmid:16935499
- 37. Wu Y, Tam NFY, Wong MH. Effects of salinity on treatment of municipal wastewater by constructed mangrove wetland microcosms. Mar Pollut Bull. 2008;57(6–12): 727–734. doi: 10.1016/j.marpolbul.2008.02.026. pmid:18374366
- 38. Jin X, Tu Q. Criterion to lake eutrophication survey. Beijing: China Environmental Science Press; 1990.
- 39. Chen GC, Tam NFY, Ye Y. Spatial and seasonal variations of atmospheric N2O and CO2 fluxes from a subtropical mangrove swamp and their relationships with soil characteristics. Soil Biol Biochem. 2012;48: 175–181.
- 40. Snedaker S, Snedaker J. The Mangrove Ecosystem: Research Methods. UK: United Nations Educational, Scientific and Cultural Organization Published, Richard Clay (The Chaucer Press) Ltd.; 1984.
- 41. Bonham CD, Ahmed J. Measurements for terrestrial vegetation. New York: John Wiley and Sons; 1989. pp. 338.
- 42. Ryden JC. Denitrification loss from a grassland soil in the field receiving different rates of nitrogen as ammonium nitrate. J Soil Sci. 1983;34(2): 355–365.
- 43. Tam N, Wong A, Wong MH, Wong YS. Mass balance of nitrogen in constructed mangrove wetlands receiving ammonium-rich wastewater: Effects of tidal regime and carbon supply. Ecol Eng. 2009;35(4): 453–462.
- 44. Tam N. Effects of wastewater discharge on microbial populations and enzyme activities in mangrove soils. Environ Pollut. 1998;102(2): 233–242.
- 45. Personeni E, Loiseau P. Species strategy and N fluxes in grassland soil—A question of root litter quality or rhizosphere activity? Eur J Agron. 2005;22(2): 217–229.
- 46. Poorter H, Vanderwerf A, Atkin OK, Lambers H. Respiratory energy-requirements of roots vary with the potential growth-rate of a plant-species. Physiol Plantarum. 1991;83(3): 469–475.
- 47. Aerts R, Chapin FS III. The mineral nutrition of wild plants revisited: a re-evaluation of processes and patterns. Adv Ecol Res. 1999;30: 1–67.
- 48. Tauchnitz N, Brumme R, Bernsdorf S, Meissner R. Nitrous oxide and methane fluxes of a pristine slope mire in the German National Park Harz Mountains. Plant Soil. 2008;303(1–2): 131–138.
- 49. Gui P, Inamori R, Matsumura M, Inamori Y. Evaluation of constructed wetlands by wastewater purification ability and greenhouse gas emissions. Water Sci Technol. 2007;56(3): 49–55. pmid:17802837
- 50. Ando T, Yoshida S, Nishiyama I. Nature of oxidizing power of rice roots. Plant Soil. 1983;72(1): 57–71.
- 51. Chiu CY, Chou CH. Oxidation in the rhizosphere of mangrove Kandelia candel seedlings. Soil Sci Plant Nutr. 1993;39(4): 725–731.
- 52. Mckee KL, Rooth JE. Where temperate meets tropical: multi-factorial effects of elevated CO2, nitrogen enrichment, and competition on a mangrove-salt marsh community. Glob Change Biol. 2008;14(5): 971–984.