Nitrogen immobilization may reduce invasibility of nutrient enriched plant community invaded by Phragmites australis

Nutrient enrichment, particularly nitrogen, is an important determinant of plant community productivity, diversity and invasibility in a wetland ecosystem. It may contribute to increasing colonization and dominance of invasive species, such as Phragmites australis, especially during wetland restoration. Providing native species a competitive advantage over invasive species, manipulating soil nutrients (nitrogen) may be an effective strategy to control the invasive species and that management tool is essential to restore the degraded ecosystems. Therefore, we examined competition between Phragmites australis and Melaleuca ericifolia in a greenhouse setting with activated carbon (AC) treatments, followed by cutting of Phragmites shoots in nutrient-rich soils. Additionally, we evaluated the effect of AC on plant-free microcosms in the laboratory, to differentiate direct effects of AC on soil microbial functions from indirect effects. Overall, the objective was to test whether lowering nitrogen might be an effective approach for reducing Phragmites invasion in the wetland. The AC reduced Phragmites total biomass more significantly in repeated cut regime (57%) of Phragmites shoots compared to uncut regime (39%). Conversely, it increased Melaleuca total biomass by 41% and 68% in uncut and repeated cut regimes, respectively. Additionally, AC decreased more total nitrogen in above-ground biomass (41 to 55%) and non-structural carbohydrate in rhizome (21 to 65%) of Phragmites, and less total nitrogen reduction in above-ground biomass (25 to 24%) of Melaleuca in repeated cut compared to uncut regime. The significant negative correlation between Phragmites and Melaleuca total biomass was observed, and noticed that Phragmites acquired less biomass comparatively than Melaleuca in AC-untreated versus AC-treated pots across the cutting frequency. AC also caused significant changes to microbial community functions across Phragmites populations, namely nitrogen mineralization, nitrification, nitrogen microbial biomass and dehydrogenase activity (P ≤ 0.05) that may potentially explain changes in plant growth competition between Phragmites and Melaleuca. The overall effects on plant growth, however, may be partially microbially mediated, which was demonstrated through soil microbial functions. Results support the idea that reducing community vulnerability to invasion through nutrient (nitrogen) manipulations by AC with reducing biomass of invasive species may provide an effective strategy for invasive species management and ecosystem restoration.

result, the nutrient enrichment in soil can benefit fast-growing invasive plant species, that outperform the native slow-growing plant species and invade into native ecosystems 8 . Wetlands, including brackish marshes, represent one of the most productive natural habitats 9 supporting various biological communities and biogeochemical functions. These provide heterogeneous nutrient conditions and diverse habitats, which can play as one of the key drivers to global nutrient cycling 9,10 . Additionally, topographic features of wetlands make these ecosystems vulnerable to human-induced disturbances 11 , thereby, facilitating invasion of non-native plants, and even some native plants play dominating role into the ecosystem.
Invasion of Phragmites australis (hereafter Phragmites) possess the great threat amongst aggressive and widely studied invasive plants in wetlands 12,13 . It significantly modifies and often destroys ecosystem structure and functions [14][15][16][17] . Physiological characteristics of Phragmites including high rates of reproduction, well-developed aerenchyma and rhizosphere facilitate colonization, and dense canopies related to their high productivity can inhibit germination and growth of neighbour plant species [18][19][20] . The tissue chemistry of Phragmites and its extensive production alters nutrient pools 21 and availability in soils 16,18 . Together with its negative effects on ecology, Phragmites can impose considerable financial loss 22 . For example, invasive species management costs about $25 billion/year in the USA 23 . Globally, there are numerous wetland restoration projects related to Phragmites invasion representing significant cost to the USA, Canada and Australia [24][25][26] . Though invasiveness is dependent upon genotypes, for example, a Eurasian genetic lineage is considered as an invader in North America 27 and Phragmites has been considered a cryptic invader creating management problems for scientists and land managers 28,29 . The invasive plant management project is generally lacking scientific information regarding invasion mechanism of the targeted species. For example, Phragmites is considered a cryptic invader, it may lack specific information related to invasion mechanisms, though it depneds on its context 30 . Current methods to control Phragmites invasions are aimed at eradication of existing populations, which are effective only in short term 31 . The recent study raises the question about the efficacy of managing large populations invaded by Phragmites through the method 22 . Generally, Phragmites populations can be treated using herbicide applications in combination with cutting, burning and covering with plastic 32 . Phragmites populations may quickly re-sprout from rhizome and seed bank reserves after the treatments 33 , indicating it requires long-term monitoring and re-treatment until native vegetation can be restored. Some recent studies showed seeds compared to vegetative propagules may play an important role to reproduce the Phragmites colonization 27 . Thus, repeating control measures indefinitely to address recurring Phragmites invasions is not only expensive, and time consuming, but also interferes with native plant community establishment 34 .
There are many studies supporting the assertion that nutrient enrichment favours invasiveness and long-term persistence of the invader in wetland native plant communities making the community more susceptible to be invasible [35][36][37] . The approach to ecosystem management like Phragmites invaded ecosystems, focus on simply eradicating Phragmites stands rather than nutrient management 38,39 and restoring native plant communities 32 . Again, very few studies have been conducted systematically on community invasibility to compare the effects of disturbance regimes 40 and how management approaches may work effectively on promoting re-establishment of neighbouring plant species 4 . Therefore, we hypothesized that creating conditions through nutrient (nitrogen) manipulation in soil followed by frequent cuttings of Phragmites shoots may favour associated plant species Melaleuca ericifolia (hereafter Melaleuca) over Phragmites that would be a more operative approach to reduce invasions by Phragmites. The Melaleuca has been selected as an alternative plant for wetland restoration because distribution and abundance of this indigenous plant has reduced sharply due to high salinity, nutrient enrichment, flooding and invasion along the entire coastal zone of Australia, especially southern-eastern Australia 39,41,42 . Thus, the rehabilitation of those wetlands is the priority of the natural resource management agencies (government and private) in Australia 43,44 . This hypothesis is associated with the resource-ratio hypothesis, that explains plant community composition influenced by relative abundance of limiting resources in soil 45,46 . Thus, the resource (nutrient) manipulation limits the availability to the target dominating plant species, but it may benefit the desired native species 47 . Our previous studies support the idea that nutrient enrichment, specifically high nitrogen (N) availability, can facilitate Phragmites invasion into wetlands and thereby, suppress the associated plant species Melaleuca 39 . In addition to N enrichment, increased phosphorus may play an important role to displace the slow-growing native plants by fast-growing invasive plants in many ecosystems including wetland ecosystems 48,49 . But, in particular, N enrichment in the wetland is the key driver to play community invasibility because a large part of N is being deposited in wetlands through surface runoff and ground water 50 . Thus, Phragmites is highly responsive to N availability and it attains a competitive advantage under these conditions 39 similar to plant species Rumex confertus 51 .
The key matter in restoration with nutrient-enriched ecosystems is to deal with the question of how the slower growing native species may benefit in competition over fast-growing invasive species 52,53 . Till now, there is no conclusive assertion about the successful mechanisms of invasive species and the susceptibility of plant communities to invasion 54,55 . Therefore, documentation of species-specific invasion mechanism, its effects and control into invaded plant communities is more reasonable and methodical than on general theories for all invasive species 54,56,57 . In addition, relatively very little research has focused on the effects of control measures of the invasive species, for example, Phragmites on the rhizome and rhizosphere soil that influence plant productivity, and likely contribute to competitive advantages over native plant species 58 . Therefore, we questioned whether depletion of rhizome reserves via shoot cutting and changes in rhizosphere soil characteristics through soil nitrogen immobilization via activated carbon (AC) amendment might be the key to control of Phragmites in its invaded communities. Activated carbon addition to soil generally reduces nitrogen availability to plants 59 . The general assumption is that AC reduces inorganic nitrogen content in soil such as NO 3 and NH 4 60 which are mostly available for plant growth and they may limit growth and productivity of plant 61 . The underlying mechanism is that addition of AC leads to nitrogen immobilization through greater microbial N uptake 62 , but it depends on soil characteristics 63 . The general understanding is that limiting N availability to plant through immobilization may reduce the www.nature.com/scientificreports www.nature.com/scientificreports/ productivity of the invasive plant, and thereby, increasing the growth of native ones, which also depends on the response of the plant to N availability in soil 64 .
Therefore, we conducted greenhouse experiments to test whether management approaches in nutrient enriched condition may increase the competitive ability of Melaleuca over Phragmites into a wetland plant community. Additionally, a laboratory soil incubation experiment was conducted to test whether AC-induced changes to soil microbial functions via nitrogen immobilization could be helpful to explain the changes in plant growth in the above experiments. Our precise objectives were to (i) study the interactive effects of AC amendment in growth substrate and repeated shoot cutting of Phragmites on the competitional growth between Phragmites and Melaleuca, and (ii) assess whether responses of plant productivity and resources (carbohydrate storage, nitrogen concentration) to AC addition and subsequent cutting of Phragmites shoots were mediated by N immobilization via microbial interactions. More specifically, the experiments were conducted to test whether AC addition followed by repeated cutting of Phragmites shoots either individually or combinedly could facilitate Melaleuca success given sufficient competition over Phragmites.

Materials and Methods
Study species. Phragmites australis is a cosmopolitan and successional species that is distributed all over the world (except Antarctica), and still it remains unclear about its originality in many regions of the earth 21 . It has been considered as a weedy and invasive population in North America, Australia, and Madagascar [65][66][67] . Phragmites is widely-dominated across most parts of temperate regions of Australia 65 . It forms mostly dense monospecific stands in fresh and coastal wetlands, freshwater swamps and lakes, along rivers, and even in irrigation canals 68 . On the other hand, Melaleuca is a native shrub or small tree with a maximum height 8 m, is distributed in coastal and near coastal freshwater and brackish-water wetlands across southeastern Australia 44 . Most species of the genus Melaleuca are reliant on seed alone for reproduction; however, Melaleuca can form clonal stands through the production of ramets 69 . These two species naturally coexist in most of the wetlands in south-eastern Australia 70,71 . Greenhouse studies. Collection of plant materials. Methods for the greenhouse experiments were followed according to our previous works 39,42 with modifications in treatments. Briefly, in September 2012, we collected rhizomes of Phragmites from Cherry Lake (37°51′30″S, 144°50′5″E), that is a coastal wetland in Altona, Melbourne, Victoria, Australia 72 . Cherry Lake was a low land and seasonally inundated marshy area before European settlement, but following urbanization, it was altered and drained to manage the wetland. During European settlement, Phragmites occupied a small portion of the wetland, but gradually, it has been expanding vigorously and making almost monospecific stands that changes the floristic composition 16 . Considered a less-disturbed and low-nutrient wetland 39 , the site was selected to source plant materials. We used the live rhizome for the experiment consisting of only one active node. We purchased 6-month old Melaleuca seedlings from a commercial source, grown from locally collected seeds in potting mix soil 42 .
Soil preparation. We used circular 7 L plastic pots (25 × 22 × 17-cm) containing 4 L substrate. We prepared the substrate with 1:7 ratio of river sand and potting mix respectively, including nutrient addition treatment (3 g/L) nutrient level. The composition of the potting mix was organic materials, living organisms, minerals and nutrients (see 'availability of materials and data' for more information). The used nutrient was the mixed pelletized fertilizer with a ratio of N-P-K: 16-8-9 and it was used to enrich nutrient level. In addition, a 50 mL of soil from mixed population of Phragmites and Melaleuca in Edithvale wetland (described below in details) was added to the substrate for introduction of microbes. This soil was also used for a later field soil incubation experiment. We mixed soil (river sand & potting mix) with inoculum and nutrients appropriately in isolated pots by hand, placing the mixture into targeted plastic pots, then used for transplantation of the excised rhizomes of Phragmites.
Experimental design and measurement. To examine the interactive effects of AC amendment in soil substrate and repeated shoot cutting of Phragmites on competition between experimental plants in nutrient enriched condition, we initially manipulated the substrate pots with AC (activated charcoal powder made from coconut shell) [with (70 g/L substrate) and without AC (0 g/L)] and nutrient (3 g/L). The manipulated pots were then subjected to nutrient addition at a rate of 3 g/L in the tilled topsoil monthly. Plants were arranged in treatment pots as Phragmites alone, Melaleuca alone, and Melaleuca with Phragmites with five replicates. The management strategies of the experimental pots were: (a) Phragmites alone and Melaleuca alone (b) Phragmites plus Melaleuca without cutting of Phragmites shoots, (c) Phragmites plus Melaleuca subjected to one time Phragmites shoot cutting at 6 weeks during the 6-month experimental period, and (d) Phragmites plus Melaleuca with two times Phragmites shoot cuttings by a 6-week interval from commencement of the experiment until termination.
We transplanted the collected live rhizomes of Phragmites and Melaleuca seedlings into prepared experimental pots as soon as possible to avoid damage to the plant materials. The experimental pots were placed in a natural lit greenhouse, with a condition of 23 ± 3 °C and 12 ± 2 °C day/night temperature. We used an auto irrigation system to water the experimental pots and maintained soil moisture at 55 ± 5%, with a weekly soil moisture monitoring system. We randomized the experimental pots weekly by cleaning other seedlings except Phragmites and Melaleuca. We applied a destructive approach for biomass measurement and collected each individual of Phragmites and Melaleuca after 6 months of their growing for measurement of above-ground biomass (AGB) and below-ground biomass (BGB). Rhizome of Phragmites was collected for total non-structural carbohydrate (TNC) content to justify the role of carbohydrate for vegetative success of Phragmites. In addition to this, total nitrogen (TN) concentration in above-ground tissue of both species was measured to differentiate the nitrogen assimilation by both of the species.
www.nature.com/scientificreports www.nature.com/scientificreports/ Laboratory incubation studies using field soil. Investigating AC-induced changes to soil nitrogen cycling rate and microbial activities [soil microbial biomass nitrogen (SMB-N) and dehydrogenase activity (DHA)] may be helpful explaining the changes in plant growth, as determined in the above experiment. The experiment was conducted in laboratory using nutrient enriched field soil and it was plant-free microcosms experiment to differentiate direct versus indirect effects of AC on the soil microbial functions. This was due to AC influencing soil effects on plant growth through allelochemicals absorption secreted by Phragmites 42,68 and altering plant growth. We collected 15 soil cores from three high nutrient enriched Phragmites invaded populations in the Edithvale wetland (38°05′54.2′′S 145°08′24.8′′E) in December 2016. The wetland is located in south-eastern Melbourne which is listed as Ramsar site on August 2001. The Edithvale wetland is a mud-flat habitat, which is considered as high nutrient enriched soil and adversely impacted by urban development 39,73 . The main source of wetland nutrient enrichment is related to urban stormwater including industrial and agricultural runoff 74 . The soil was considered for this soil incubation experiment due to its high nutrient-enriched characteristics. The collected soil amended with and without AC was examined to compare how AC may change soil nitrogen cycling and microbial activities to understand the roles of nitrogen immobilization and microbial functions in controlling the competitiveness between Phragmites and Melaleuca in the above greenhouse studies.
The five invaded cores from each of the considered Phragmites population were collected and each individual core was kept in separate Ziploc bags. The samples were immediately transported to the laboratory with a laboratory ice box. Soils of the five replicates in each population were homogenised by sorting all debris and sieving (<2 mm). Before treatment, five subsamples of soil (10 g each) were collected from each composite and immediately extracted with 50 mL of 2.0 M KCl for the initial determination of extractable ammonium and nitrate. To half of the soil from each population, 70 g/L AC (the amount used for above greenhouse studies) was mixed into the soil. All soils with and without AC were rearranged in 5 replicates into the new plastic soil pot and maintained 50% soil moisture by tap water that was compatible with the field water-holding capacity. Then, the plastic soil pots were kept in dark at 22 °C for 30 days to simulate microbial activities and functions. After 30 days, the soils were removed, and subsamples were taken for specific laboratory analyses.
Laboratory analysis. We used ion chromatograph to measure ammonium and nitrate concentrations in all soil extracts (Shimadzu Ion Chromotograph, Kyoto, Japan) following the method of Maynard et al. 75 . Net mineralization of N was calculated as extractable nitrate (NO 3 − ) + ammonium in the incubated sample minus extractable nitrate + ammonium in the initial extracts. Net nitrification was calculated as extractable nitrate in the incubated sample minus extractable nitrate in the initial extracts. Microbial biomass nitrogen (MBN) and dehydrogenase activity (DHA) in treated soil were determined according to the procedures of Uddin & Robinson 71 . The alkaline persulfate oxidation method was used to determine the total nitrogen in above-ground plant tissue of the above greenhouse experimental samples of both species following the method proposed by Purcell & King 76 . We used phenol-sulphuric acid method for TNC measurement in Phragmites rhizome of the greenhouse experiment 77 .

Statistical analyses.
We applied analysis of variance (ANOVA) and nested ANOVA to assess the effects Phragmites, AC and Phragmites shoots cuttings on the measured variables of Melaleuca. In the case of Melaleuca, the model comprised of main effects of Phragmites & AC (effect of one of the independent variables on the dependent variables), and their interactions. In addition, the effects of cuttings of Phragmites shoots (nested within Phragmites), the cutting frequency of Phragmites shoots (nested within cutting), and their interactions with AC also formed the part of the model, which was analysed by the statistical methods of Uddin et al. 42 . In case of Phragmites, we used ANOVA models that included only cutting treatments of Phragmites, AC and their interactions. One-way ANOVA with LSD tests were used to compare the multiple means among treatments for different measured parameters of Phragmites and Melaleuca. We also used ANOVAs to examine the effects of AC addition on soil extractable nitrate and ammonium concentration, net mineralization and nitrification, microbial biomass-N, and DHA in the field soil incubation experiment. The model used included Phragmites populations and AC, and their interactions. Again, we used Real Statistics Resource Pack (Software plug-in, Microsoft Excel 2010) to investigate if the difference in slopes of regressions between AC treatments was statistically significant 78 . Furthermore, independent sample t-test was conducted to compare soil nitrogen indices within each population. We also applied Levene's test of equality to check the homogeneity of variance of the used data (square root transformed if necessary) for both of Phragmites and Melaleuca. We considered P ≤ 0.05 as significant for all experimental data analyses by using statistical software IBM SPSS statistics 24.0.

Results
plant responses to activated carbon and shoot cuttings of Phragmites. The aboveground and belowground biomass of Melaleuca was negatively affected significantly by competition with Phragmites (Table 1, Fig. 1A,B). The average Melaleuca AGB decreased from 692.24 ± 37.38 to 32.29 ± 1.03 g/m 2 and 6399.55 ± 34.09 to 45.02 ± 2.06 g/m 2 between mono (only Melaleuca) and mixed uncut cultures (Phragmites and Melaleuca grown together), either AC-untreated or AC-treated, respectively (Fig. 1A,B). In case of BGB, it reduced from 205.02 ± 16.55 to 9.92 ± 0.44 g/m 2 and 180.94 ± 9.74 to 14.61 ± 0.90 g/m 2 respectively (Fig. 1A,B). AC had no significant growth effect on Melaleuca in mono-culture whereas it had significant effect on mixed uncut cultures (Fig. 1A,B). Again, AC had significant negative growth effect on Phragmites in both mono and mixed uncut cultures ( Table 2, Fig. 2A,B). The AGB and BGB of Phragmites was reduced from 1979.88 ± 79.81 to 1290.37 ± 50.95 g/m 2 and 2311.22 ± 95.40 to 1660.33 ± 77.55 g/m 2 between mono and mixed uncut cultures without and with AC, respectively. The addition of AC had different outcomes on the competitive interactions between Phragmites and Melaleuca biomass, but it was noticed that Melaleuca growth increased due to AC in uncut regime ( Figs. 1 and 2). The interactive effects of AC and Phragmites were significant effect on biomass of Melaleuca (Table 1).
The cutting had significant negative effect on biomass of Phragmites, and consequently, it increased biomass of Melaleuca (Tables 1 and 2  www.nature.com/scientificreports www.nature.com/scientificreports/ Total nitrogen concentration in above-ground biomass of Phragmites was greater than Melaleuca (Fig. 3A,B). TN concentration of Melaleuca was negatively affected significantly with Phragmites competition and AC treatments (Table 1). Cutting and cutting frequency of Phragmites shoots also significantly increased the TN concentration of Melaleuca among mixed cultures, whereas the TN concentration of Phragmites was decreased with both AC-untreated and AC-treated cultures (Fig. 3A,B), but however, no significant interactions between AC and cutting regimes were found ( Table 1). The average Melaleuca TN concentration was reduced from 31.72 ± 0.90 to 8.17 ± 0.42 mg/g with competition of Phragmites without AC, which was 74%, whereas with AC it was 72% (22.20 ± 0.58 to 6.09 ± 0.28 mg/g). Again, cutting of the Phragmites shoots has significant positive effect in increasing the TN concentration of Melaleuca, which was an average from 8.17 ± 0.42 to 13.77 ± 0.43 mg/g without AC and 6.09 ± 0.28 to 10.06 ± 0.42 mg/g with AC, whereas TN concentration of Phragmites was reduced from 40.55 ± 0.71 to 24.59 ± 0.48 mg/g without AC and 23.90 ± 0.49 to 11.19 ± 0.74 mg/g with AC.   www.nature.com/scientificreports www.nature.com/scientificreports/ The concentrations of TNC in Phragmites rhizomes strongly differed between the treatments (Table 2, Fig. 4). The imposed management regimes affected the levels of TNC, and TNC concentrations were significantly higher at without AC than with AC. Single shoot cuttings of Phragmites reduced 32% (111.88 ± 3.05 to 76.46 ± 3.03 mg/g) & 46% (88.19 ± 2.75 to 47.41 ± 2.62 mg/g) TNC concentration in without and with AC respectively, whereas they were 53% & 79% for double shoot cuttings (Fig. 4). However, cutting of Phragmites shoot had significant negative effects on TNC at both levels of AC. There were no competition effects on TNC at both AC levels ( Table 2), but  www.nature.com/scientificreports www.nature.com/scientificreports/ significant interactions between AC and cutting management were found for TNC indicates that the effects of these management regimes were effective ( Table 2).

Relationships among measured variables. There was significant negative correlation between
Phragmites and Melaleuca total plant biomass in all experimental pots (Fig. 5A). It was observed that Phragmites acquired greater biomass in AC-untreated pots compared to AC-treated pots across the cutting frequency (Fig. 5A). In contrast, Melaleuca showed the opposite scenario regarding biomass accumulation. The significant difference was found between slopes of the regression lines of AC-untreated and AC-treated pots (t = 3.47) with df 26 and α 0.05, and the stronger negative correlation was noticed in AC-untreated pots (F = 57.12, df = 1,12, p > 0.001; r 2 = 0.85) compared to AC-treated pots (F = 36.55, df = 1,12, p > 0.001; r 2 = 0.76) (Fig. 5A). Again, a significant positive relationship between aboveground biomass and total nitrogen concentration of Melaleuca and Phragmites in mixed culture both in AC-untreated and AC-treated was found, but the responses were different (Fig. 5B,C). The significant difference was found between slopes of the regression lines of AC-untreated and AC-treated pots for Melaleuca (t = 3.38) (Fig. 5B), but not for Phragmites (t = 0.5) with df 26 and α 0.05 (Fig. 5C) (Fig. 5B,C). Furthermore, it was found that there was a significant positive correlation between below-ground biomass of Phragmites and total non-structural carbohydrates in both AC-untreated and AC-treated mixed communities (Fig. 5D). The slope difference between AC-untreated and AC-treated was significant (t = 2.69) with df 26 and α 0.05 that indicates AC had a significant effect in influencing the carbohydrate storage in rhizomes (Fig. 5D). The stronger positive relation was noted in AC-treated pots (F = 131.17, df = 1,12, p > 0.001; r 2 = 0.91) compared to AC-untreated pots (F = 75.77, df = 1,12, p > 0.001; r 2 = 0.86) (Fig. 5D). Again, the strong positive relationship between TN concentration of AGB and TNC level of BGB in Phragmites was noticed in AC-untreated (F = 67.36, df = 1,12, p > 0.001; r 2 = 0.85) compared to AC-treated (F = 49.97, df = 1,12, p > 0.001; r 2 = 0.82) mixed communities (Fig. 5E), but the slope difference was not significant (t = 1.85) with df 26 and α 0.05.

Soil responses to activated carbon. Concentrations of extractable NO 3 -N and NH 4 -N were greatly
reduced in pots treated with AC relative to unamended pots at the end of the experiment in all populations ( Table 3, Fig. 6A,B). Soils amended with AC reduced an average 19% and 17% NO 3 −N and NH 4 -N respectively. The reductions of NO 3 -N were from 66.07 ± 1.03 to 54.72 ± 1.59 mg/kg, 57.09 ± 1.21 to 45.89 ± 1.00 mg/kg and 58.28 ± 1.97 to 47.45 ± 1.82 mg/kg in population_1, population_2 and population_3 respectively (Fig. 6A). The NH 4 -N was reduced by an order of scale, from 12.24 ± 0.45 to 9.94 ± 0.34 mg/kg, 9.77 ± 0.36 to 7.70 ± 0.17 mg/ kg, 10.01 ± 0.55 to 8.57 ± 0.10 mg/kg in population_1, population_2 and population_3 respectively (Fig. 6B). AC addition decreased both net N mineralization and net N nitrification significantly, which was on an average 77% and 73% respectively compared to without AC pots in all populations (Table 3, Fig. 6C,D), but Phragmites population reduced significantly only net N mineralization (Table 3). Microbial biomass N was significantly higher in AC unamended pots than amended pots, which were 40.81, 38.59 & 44.28% increment in popula-tion_1, population_2 and population_3, respectively (Table 3, Fig. 7A). However, the microbial biomass N was significantly influenced by Phragmites populations, but there is no significant interaction between AC and populations ( Table 3). The microbial biomass N decreased on an average 41% at AC-amended soil compared to AC unamended soil. Dehydrogenase activity was also significantly lower in the AC-amended soil by an average 43.57% than in the AC-unamended soil (Fig. 7B). AC-amendment decreased dehydrogenase activity most significantly in population_2 among three populations (Fig. 7B). The overall reduction of dehydrogenase activity indicated a decrease in the microbial metabolism in the soil due to less mineralization and nitrification rate in the AC-amended soil ( Table 3, Fig. 6C,D). All parameters except net nitrification in soil significantly varied across the AC-amendment and populations but no interactive effects (AC × Phragmites populations) were observed (Table 3).

Discussion
Plant responses. In nutrient-enriched wetlands, Phragmites seedlings and adult plants grow larger and produce more stems, florets, inflorescences and thus, seeds with rhizome, further benefiting the invasion 79 . We hypothesized that reducing nitrogen availability in nutrient-rich soil via AC addition followed by Phragmites shoot cuttings may reduce Phragmites growth, and thereby, may favour growth of associate plant species Melaleuca. So, controlling growth of Phragmites in nutrient-rich wetlands, our experimental results demonstrate that AC-amended soil followed by Phragmites shoot cuttings increased competitiveness of Melaleuca over Phragmites growth (Fig. 1). On the contrary, Phragmites competitively overwhelmed Melaleuca growth in AC unamended soil (Fig. 1), which is aligned with other studies in prairie restoration 80 . The AC additions in our studies reduced biomass of Phragmites that indicates the reduction of invasibility of the plant community invaded by Phragmites. Conversely, AC increased biomass of Melaleuca in mixed culture, which suggest that AC had negative impact on Phragmites allowing Melaleuca to be more competitive.
Our results are largely consistent with the fluctuating resource hypothesis of invasibility proposed by Davis et al. 81 , whom suggests that a plant community is more susceptible to invasion when there is an increase in the amount of unused resources. This condition also increases the relative competitive ability of invasive species in a plant community 82 . In our experimental results, the increase of Melaleuca biomass in AC-amended mixed cultures indicated a relatively low positive outcome because Melaleuca is a tree with lengthy lifespan, it requires longer time to achieve more biomass than Phragmites, a perennial grass. Melaleuca had also better growth (shown in single growth experiment) suggesting that with the correct growth condition, it would successfully establish Scientific RepoRtS | (2020) 10:1601 | https://doi.org/10.1038/s41598-020-58523-4 www.nature.com/scientificreports www.nature.com/scientificreports/ www.nature.com/scientificreports www.nature.com/scientificreports/ in its community. As a result, it can be said Melaleuca may establish and grow properly in Phragmites-dominated wetland ecosystems with AC-amendment followed by reducing biomass of Phragmites.
The success of the establishment of plants depends on mechanisms, and the mechanisms are species-specific and contextual 83 . It is difficult to establish a single management approach to an ecological restoration initiative, requiring integrated control measures to restore the degraded ecosystem. This would be practical if the targeted plant, for example, Phragmites has persistent communities 84 . In our studies, however, only the AC addition reduced Phragmites biomass and increased Melaleuca biomass, but AC with cutting of Phragmites shoots enhanced the magnitude of reduction for Phragmites biomass. Subsequently, this approach increased the biomass of Melaleuca more significantly suggesting combined management (AC with cuttings of shoots) was more effective than only AC addition to reduce the invasibility of the community by Phragmites. Melaleuca experienced comparatively less impact by Phragmites in AC-amended pots compared with AC-unamended cutting treatments  Table 3. Results of ANOVA analysing treatments of activated carbon, Phragmites population, and their interactions on soil nitrate (NO 3 − N), ammonium (NH 4 + -N), microbial biomass nitrogen (MB-N), and dehydrogenase activity (DHA) in nutrient enriched Phragmites invaded soil. a AC = activated carbon, PP = Phragmites population. www.nature.com/scientificreports www.nature.com/scientificreports/ pots (Fig. 1). It may imply that AC-induced low N availability in soil had less effect on the competitive ability of Melaleuca and Phragmites. It is assumed that Melaleuca achieved benefits to establishment and growth in cutting experiments with AC-amendment through nitrogen immobilization, which was also effective in reducing Phragmites competitive ability. Thus, our results were compatible with findings of Perry et al. 85 whom found lowering N availability through carbon amendment reduced the competitive ability of invasive species, Phalaris arundinacea. As a result, we can assume that the effects of AC addition with repetitive cutting of Phragmites shoot on Melaleuca growth is explained through N immobilization.
The increasing trend of total N concentration in AGB of Melaleuca and decreasing level in Phragmites AGB among treatments indicates that AC associated with cuttings of Phragmites shoots offered competitive ability of Melaleuca by up-taking more available nitrogen over Phragmites. This is aligned with other studies, that found invasive species attained comparatively less nitrogen content than native in the used treatments 85,86 . We also examined the effects of treatments on carbohydrates reserves in Phragmites rhizome that allow Phragmites to re-sprout in the next spring growth. The treatments reduced significantly Phragmites biomass (above and belowground) and carbohydrate reserves in rhizome. Additionally, our laboratory incubation experiment showed that AC also had direct impact on soil N availability. This assumes that it might be a basic function of both N-availability in soil and carbohydrate reserve in rhizome for competitive advantage of Phragmites in the plant community. This implies Phragmites establishment and growth might be restricted by carbohydrate reserve in rhizome and N-availability in soil. This finding is aligned with the studies of Kleijn et al. 87 , whom found available soil N supplemented by stored carbohydrate reserves was responsible for vegetative success of Veratrum album.
The negative correlation between Phragmites and Melaleuca total biomass (Fig. 5A), and the more reduction of Melaleuca biomass in AC-untreated compared to AC-treated experimental pots (Fig. 1A,B) indicates that AC with cutting of Phragmites shoots may promote the productivity of Melaleuca, which is aligned other studies 62 . The studies found carbon amendment reduced N availability in soil, and thereby, reduced competition between native and exotic plant, and provided benefit to the native. The more positive relationship between nitrogen concentration and above-ground biomass in Phragmites (Fig. 5B,C) indicates responses of Phragmites to N availability was greater than Melaleuca, which may provide it indirect benefit. Melaleuca however, may experience less negative effects than Phragmites for the reduction of N availability, which is supported by the studies of Alpert and Maron 88 and Alpert 64 . They found carbon addition may reduce ecosystem invasiblity by invasive grass through lowering nitrogen availability and promoting native plants growth for ecosystem restoration. www.nature.com/scientificreports www.nature.com/scientificreports/ Furthermore, AC treatment with cuttings influence TNC levels of rhizomes indicates that the management options may work in combination to slowly reduce carbohydrate reserves of Phragmites rhizome, thus ultimately negatively influencing the sprouting for the next growth. The results are aligned with the studies of Moyo et al. 89 , whom found cutting management had significant negative effect on TNC level. The reducing level was amplified with cutting frequency that ultimately affected shoot production. Additionally, the positive relationships among rhizome biomass, TNC of Phragmites rhizome and TN concentration of Phragmites AGB (Fig. 5D,E) indicate that reducing rhizome biomass through cutting and AC-amendment may act as a negative driver to re-sprout for Phragmites, which ultimately provides benefits to Melaleuca. The findings are similar with other studies of Druege et al. 90 whom found nitrogen concentration and TNC level are positively correlated, and cutting frequency may reduce TNC level in Chrysanthemum plant species.
Soil responses. In our study, extractable inorganic N concentrations (nitrate and ammonium concentrations) were lower in AC pots relative to unamended pots, consistent with the studies of Blumenthal et al. 59 and Morghan et al. 91 , where addition of AC as sucrose and/ or sawdust has shown reduced soil inorganic N levels, though the efficacy depends on initial soil fertility, quantity and form of AC added. Again, our results showed that AC additions stimulated microbial N-immobilization by reducing rates of net N mineralization and net nitrification, which is consistent with the findings of Torok et al. 92 , who found nitrogen immobilization by carbon acts to stimulate the restoration process in a grassland community. The results of our laboratory incubation are compatible to explain the underlying mechanism of nitrogen cycling in the first plant community experiment as the N level in the used field soil for incubation experiment was high 39 .
The MBN significantly decreased in the AC-amended soil in all populations in our study, which is supported by other studies 93,94 . For example, Zhang et al. 93 found the MBN decreased with AC addition at a rate of 4.5 t/ ha/yr in the field at most sampling dates and soil depths compared to the unamended soil treatment, whereas Zavalloni et al. 95 and Dempster et al. 96 found AC amendment had no significant impact on MBN. Again, AC increased soil MBC in our studies supported by Ma et al. 97 and Gebhardt et al. 98 whom found addition of carbon as a soil amendment significantly increased MBC and thereby, altered plant biomass. Therefore, the increased MBC and decreased MBN indicates that AC in soil acted as a carbon source rather than a nitrogen source for microbes in soil substrates. As a result, this had a significant influence on N cycling, for example, increasing microbial N immobilization, which is also aligned with our results. AC is not only responsible for reducing soil N level, but it may work in soil through different mechanisms. For example, some studies assumed that AC changes plant growth by sequestering allelochemicals, that was found in previous Phragmites studies including ours 42 , and altering microbial functions through changes of microbial communities 99,100 . In this case, soil DHA may be used as an indicator of overall soil microbial activity 101 . In our study, AC caused direct changes to soil microbial functions, which were observed in our plant-free microcosms laboratory incubation experiment. Our studies showed AC decreased DHA compared to unamended soil for each population, which might be linked to decreasing the microbial symbionts associated with Phragmites success, which is explained by Centaurea species to support this argument 102,103 . Mechanisms involved in reducing invasibility. It is common for soil-N availability to influence positively for rapid shoot growth in plant and ultimately, increasing plant biomass 39,87 . In AC-amended soil, Melaleuca attained relatively higher biomass and N level with increasing cutting frequency of Phragmites shoots, whereas Phragmites experienced a lower level. This implies N-poor soil might be a good condition for the competitive ability of Melaleuca. It also indicates that N-use efficiency of Melaleuca in AC-amended soil would be higher than Phragmites, which was also supplemented by the Phragmites shoot cuttings. The findings of our studies are compatible with Perry et al. 85 , whom found native plant Carex hystericina achieved greater N-uptake and biomass compared to invasive species Phalaris arundinacea in carbon-treated soil. Again, Chapin et al. 104 found both nutrient level in soil and carbohydrate reserves play an important role to establishment of the plant, but the rapid growth is regulated completely by carbohydrate reserves. In this case, our results support the idea that reduced carbohydrate and nitrogen level in AC-amendment followed by Phragmites shoot cuttings influenced negatively the competitive ability and plant growth of Phragmites, whereas it allowed Melaleuca the opportunity to grow and establish.
Furthermore, it is assumed that N availability in soil and carbohydrate reserve in rhizome are responsible in facilitating Phragmites invasion and to completing its growing season in the short term. In our study, it appeared that while shoot cutting with AC treatments significantly reduced carbohydrates simultaneously of Phragmites, the remaining carbohydrate reserves may act as a survivor for the next spring growth. This was noted in another study related to perennial Veratrum album by Schaffner et al. 105 . So, it is important to continue the process until next growing season of Phragmites, but it is subject to context to making the approach effective in control of Phragmites in nutrient enriched condition.
implications for Phragmites control. Melaleuca growth improvement over Phragmites in AC-amended soil followed by cutting treatments suggests that this condition lowered soil N availability through N-immobilization that reduced Phragmites growth. This situation is shown between Phragmites and Melaleuca interactions, and requires further research to demonstrate the interactions between other associated plant species and Phragmites in wetlands within this condition. Creating N-immobilization in Phragmites dominated wetland restoration technique needs active reduction of soil N availability by using AC. Additionally, N availability in wetlands may be reduced by frequent cutting of Phragmites shoots before starting a restoration initiative, which supports the studies of Verhoeven & Schmitz 106 , whom found frequent mowing was effective in lowering the nutrient condition, that controls the plant growth 8 . Nitrogen immobilization technique via activated carbon has potential to reduce invasibility and provide positive effect on native plant growth in ecological restoration initiatives 107,108 .
The challenge is however, that this restoration technique needs higher amount of AC for larger scale implications, thus entailing higher cost. Maintaining the reduced N-availability in soil presents additional challenge, as many environmental processes (mineralization) and factors (surface runoff due to agriculture & industrialization) actively work to increase the N-inputs in wetlands. For this reason, it is essential to develop a new technique to reduce nitrogen level in Phragmites dominated wetland, which is partially supplemented by frequent shoot cutting of Phragmites. The minimum level of AC required for lowering N level in wetland soil prior to restoration initiative also needs to be determined, though it is contextual. Furthermore, it is essential to prevent the N-source to Phragmites dominated wetlands, which can be manipulated through vegetation buffering and landscaping 109 . Further study is recommended to investigate the wide range of N-assimilation between Melaleuca and Phragmites under this condition to find out the difference amongst them.

conclusion
Our past studies 39 showed Phragmites is more dominating in nutrient-rich compared to nutrient-poor condition which provides further evidence in supporting the hypothesis that nutrient enrichment facilitates invasion of Phragmites. Based on this scenario, this study demonstrated whether nutrient (nitrogen) manipulation may provide a competitive advantage to Melaleuca over Phragmites that might be an operative and effective restoration tool to restore the Phragmites-dominated wetlands. Our results suggest that AC addition in soil followed by frequent Phragmites shoot cuttings may provide competitive advantage for Melaleuca over Phragmites. Additionally, it is assumed that soil nitrogen immobilization due to AC addition may be responsible for Phragmites growth reduction leading to facilitating Melaleuca growth. However, soil inorganic nitrogen availability was not measured in the plant microcosms experiment, but it was addressed by plant-free microcosms and supported by the studies of Perry et al. 85 whom found inorganic nitrogen was reduced in carbon-enriched soil by 10 to 30 mg/kg in a competitive experiment between invasive Phalaris arundinacea and a native sedge Carex hystericina. Furthermore, Torok et al. 92,110 found immobilization of soil nitrogen by carbon addition was effective for the sandy grassland restoration. To the best of our knowledge, these types of experimental studies regarding Phragmites management are the first where nutrient (nitrogen) manipulation through AC followed by Phragmites shoot cuttings have demonstrated the examination of competitive ability between Phragmites and associated native species, further explained by microbial interactions that may mediate the growth of native over dominating species Phragmites.

Data availability
Data will be available where applicable.