Revealing sulfur-iron coupling mechanism for enhanced autotrophic denitrification in ecological floating beds.

A sulfur-iron coupled ecological floating bed (EFB-SFe) was developed to enhance the denitrification capability of sulfur-based ecological floating beds (EFB-S). The denitrification performance, kinetic process and microbial community composition were explored. Results showed that sulfur-iron coupling effectively enhanced the denitrification performance of EFB, surpassing the sum of their individual effects. The average total nitrogen removal rate ranged from 1.56 to 4.56 g·m-2·d-1, with a removal efficiency of 22-84 %. The k value for the S + Fe group increased from 0.04 to 0.18 d-1 to 0.40-0.46 d-1 relative to the S group. The sulfur-iron coupling promoted the enrichment of denitrifying bacteria (Thiobacillus and Ferritrophicum). The denitrification genes in EFB-SFe were upregulated, being 12-22 times more abundant than in EFB-S. Sulfur and iron autotrophic denitrification were identified as the main nitrogen removal processes in EFB-SFe. Overall, sulfur-iron coupling showed the potential to enhance the denitrification capacity of EFB-S for treating low-pollution water.

Boosting the denitrification efficiency of iron-based constructed wetlands in-situ via plant biomass-derived biochar: Intensified iron redox cycle and microbial responses.

Considering the unsatisfied denitrification performance of carbon-limited wastewater in iron-based constructed wetlands (ICWs) caused by low electron transfer efficiency of iron substrates, utilization of plant-based conductive materials in-situ for improving the long-term reactivity of iron substrates was proposed to boost the Fe (III)/Fe (II) redox cycle thus enhance the nitrogen elimination. Here, we investigated the effects of withered Iris Pseudacorus biomass and its derived biochar on nitrogen removal for 165 days in ICWs. Results revealed that accumulate TN removal capacity in biochar-added ICW (BC-ICW) increased by 14.7 % compared to biomass-added ICW (BM-ICW), which was mainly attributed to the synergistic strengthening of iron scraps and biochar. The denitrification efficiency of BM-ICW improved by 11.6 % compared to ICWs, while its removal capacity declined with biomass consumption. Autotrophic and heterotrophic denitrifiers were enriched in BM-ICW and BC-ICW, especially biochar increased the abundance of electroactive species (Geobacter and Shewanella, etc.). An active iron cycle exhibited in BC-ICW, which can be confirmed by the presence of more liable iron minerals on iron scraps surface, the lowest Fe (III)/Fe (II) ratio (0.51), and the improved proportions of iron cycling genes (feoABC, korB, fhuF, TC.FEV.OM, etc.). The nitrate removal efficiency was positively correlated with the nitrogen, iron metabolism functional genes and the electron transfer capacity (ETC) of carbon materials (P < 0.05), indicating that redox-active carbon materials addition improved the iron scraps bioavailability by promoting electron transfer, thus enhancing the autotrophic nitrogen removal. Our findings provided a green perspective to better understand the redox properties of plant-based carbon materials in ICWs for deep bioremediation in-situ.

Optimization of microplastic removal based on the complementarity of constructed wetland and microalgal-based system.

As one of the emblematic emerging contaminants, microplastics (MPs) have aroused great public concern. Nevertheless, the global community still insufficiently acknowledges the ecological health risks and resolution strategies of MP pollution. As the nature-based biotechnologies, the constructed wetland (CW) and microalgal-based system (MBS) have been applied in exploring the removal of MPs recently. This review separately presents the removal research (mechanism, interactions, implications, and technical defects) of MPs by a single method of CWs or MBS. But one thing with certitude is that the exclusive usage of these techniques to combat MPs has non-negligible and formidable challenges. The negative impacts of MP accumulation on CWs involve toxicity to macrophytes, substrates blocking, and nitrogen-removing performance inhibition. While MPs restrict MBS practical application by making troubles for separation difficulties of microalgal-based aggregations from effluent. Hence the combined strategy of microalgal-assisted CWs is proposed based on the complementarity of biotechnologies, in an attempt to expand the removing size range of MPs, create more biodegradable conditions and improve the effluent quality. Our work evaluates and forecasts the potential of integrating combination for strengthening micro-polluted wastewater treatment, completing the synergistic removal of MP-based co-pollutants and achieving long-term stability and sustainability, which is expected to provide new insights into MP pollution regulation and control.

Effects of ferrous addition to Vallisneria natans: An attempt to apply ferrous to submerged macrophyte restoration.

Restoration of submerged macrophytes is an efficient way for endogenous nutrient control and aquatic ecological restoration, but slow growth and limited reproduction of submerged macrophytes still exist. In this research, the effect of ferrous on the seed germination and growth of Vallisneria natans (V. natans) were studied through aquatic simulation experiments and its influence on the rhizosphere microbial community was also explored. The seed germination, growth, and physiological and ecological parameters of V. natans were significantly affected by the ferrous treatments. Ferrous concentration above 5.0 mg/L showed significant inhibition of seed germination of V. natans and the best concentration for germination was 0.5 mg/L. During the growth of V. natans, after ferrous was added, a brief period of stress occurred, which then promoted the growth lasting for about 19 days under one addition. The diversity and richness of the rhizospheric microbial were increased after the ferrous addition. However, the function of the rhizospheric microbial community showed no significant difference between different concentrations of ferrous adding in the overlying water. Ferrous addition affected the growth condition of plants (content of CAT, Chl a, Chl b, etc.), thus indirectly affecting the rhizospheric microbial community of V. natans. These impacts on V. natans and rhizosphere microorganisms could generalize to other submerged macrophytes in freshwater ecosystems, particularly which have similar habits. These findings would contribute to the ecological evaluation of ferrous addition or iron-containing water, and provide a reference for submerged macrophytes restoration and ecological restoration in freshwater ecosystems.

Role of hydrophytes in constructed wetlands for nitrogen removal and greenhouse gases reduction.

With the prominence of global climate change and proposal of carbon reduction concept, how to maximize the comprehensive effect of nitrogen removal and greenhouse gases (GHGs) reduction in constructed wetlands (CWs) has become crucial. As indispensable biological component of CWs, hydrophytes have received extensive attention owing to their application potential. This review comprehensively evaluates the functions of hydrophytes in nitrogen removal and GHGs reduction in CWs in terms of plants themselves, plant-mediated microbes and plant residues (hydrophyte carbon sources and hydrophyte-derived biochars). On this basis, the strategies for constructing an ideal CW system are put forward from the perspective of full life-cycle utilization of hydrophytes. Finally, considering the variability of plant species composition in CWs, outlooks for future research are specifically proposed. This review provides guidance and novel perspectives for the full life-cycle utilization of hydrophytes in CWs, as well as for the construction of an ideal CW system.

Efficient nitrogen removal pathways and corresponding microbial evidence in tidal flow constructed wetlands for saline water treatment.

The artificial tidal wetlands ecosystem was believed to be a useful device in treating saline water, and it played a significant part in global nitrogen cycles. However, limited information is available on nitrogen-cycling pathways and related contributions to nitrogen loss in tidal flow constructed wetlands (TF-CWs) for saline water treatment. This study operated seven experimental tidal flow constructed wetlands to remove nitrogen from saline water at salinities of 0-30‰. Stable and high NH4+-N removal efficiency (∼90.3%) was achieved, compared to 4.8-93.4% and 23.5-88.4% for nitrate and total nitrogen (TN), respectively. Microbial analyses revealed the simultaneous occurrence of anaerobic ammonium oxidation (anammox), dissimilatory nitrate reduction to ammonium (DNRA), nitrification and denitrification, contributing to nitrogen (N) loss from the mesocosms. The absolute abundances were 5.54 × 103-8.35 × 107 (nitrogen functional genes) and 5.21 × 107-7.99 × 109 copies/g (16S rRNA), while the related genera abundances ranged from 1.81% to 10.47% (nitrate reduction) and from 0.29% to 0.97% (nitrification), respectively. Quantitative response relationships showed ammonium transformation were controlled by nxrA, hzsB and amoA, and nitrate removal by nxrA, nosZ and narG. Collectively, TN transformation were determined by narG, nosZ, qnorB, nirS and hzsB through denitrification and anammox pathways. The proportion of nitrogen assimilation by plants was 6.9-23.4%. In summary, these findings would advance our understanding of quantitative molecular mechanisms in TF-CW mesocosms for treating nitrogen pollution that caused algal blooms in estuarine/coastal ecosystems worldwide.

Microbial response to nitrogen removal driven by combined iron and biomass in subsurface flow constructed wetlands with plants of different ages.

This study investigated the nitrogen removal enhanced by combined iron scraps and plant biomass, and its microbial response in the wetland with different plant ages and temperatures. The results showed that older plants benefitted the efficiency and stability of nitrogen removal, which could reach 1.97 ± 0.25 g m-2 d-1 in summer and 0.42 ± 0.12 g m-2 d-1 in winter. Plant age and temperature were the main factors determining the microbial community structure. Compared with temperature, plant ages affected more significantly on relative abundance of microorganisms such as Chloroflexi, Nitrospirae, Bacteroidetes and Cyanobacteria, and functional genera for nitrification (e.g., Nitrospira) and iron reduction (e.g., Geothrix). The absolute abundance of total bacterial 16S rRNA ranged from 5.22 × 108 to 2.63 × 109 copies g-1 and presented extremely significant negative correlation to plant age, which would lead to a decline in microbial function on information storage and processing. The quantitative relationship further revealed that the ammonia removal was related to 16S rRNA and AOB amoA, while nitrate removal was controlled by 16S rRNA, narG, norB and AOA amoA jointly. These findings suggested that a mature wetland for nitrogen removal enhancement should focus on aging microbes caused by old plants and possible endogenous pollution.

Salinity effect on denitrification efficiency with reed biomass addition in salt marsh wetlands.

Five simulated salt marsh wetlands with reed were constructed to investigate the effect of salinity on denitrification efficiency and its enhancement by reed biomass addition. It was found that the salinity of 7 ‰ and 10 ‰ could promote the organic carbon release of reed biomass. Results showed that the nitrate removal was highest at the salinity of 7 ‰, and would be further enhanced from 54.06 ± 12.46 % to 74.37 ± 11.53 % after the addition of reed biomass. Meanwhile, the lowest nitrous oxide emission flux was also achieved, with 0.23 mg/(m2 h) at this salinity. Microbiological analysis showed that salinity changed the microbial community. The increasing salinity increased the relative abundance of Chloroflexi and Actinobacteria, but decreased that of Proteobacteria. Main functional genera of denitrification changed from Desulfuromonas to Azoarcus and Anaeromyxbacter when the salinity increased to 15 ‰. These results will help to understand the nitrogen removal capacity of salt marsh wetlands with reed biomass addition.

New insight and enhancement mechanisms for Feammox process by electron shuttles in wastewater treatment - A systematic review.

Ammonium oxidation coupled to Fe(III) reduction (Feammox) is a newly discovered iron-nitrogen cycle process of microbial catalyzed NH4+ oxidation coupled with iron reduction. Fe(III) often exists in the form of insoluble iron minerals resulting in reduced microbial availability and low efficiency of Feammox. Electron shuttles(ESs) can be reversibly oxidized and reduced which has the potential to improve Feammox efficiency. This review summarizes the discovery process, electron transfer mechanism, influencing factors and driven microorganisms of Feammox, ang expounds the possibility and potential mechanism of ESs to enhance Feammox efficiency. Based on an in-depth analysis of the current research situation of Feammox for nitrogen removal, the knowledge gaps and future research directions including how to apply ESs enhanced Feammox to promote nitrogen removal in practical wastewater treatment have been highlighted. This review can provide new ideas for the engineering application research of Feammox and strong theoretical support for its development.

Mixotrophic denitrification improvement in ecological floating bed: Interaction between iron scraps and plant biomass.

In this study, an iron scrap (IS)-based ecological floating bed was constructed to couple with plant biomass (FeB-EFB) for treating low-polluted water, and the nitrogen removal performance and mechanism were explored. The results showed that the nitrogen could be effectively removed in FeB-EFB, and the nitrate removal efficiency was 29.14 ± 8.06% even at a low temperature (13.9 ± 2.2 °C). After the temperature rose to 20.0 ± 0.9 °C, the denitrification rate was increased by 0.63 ± 0.16-0.81 ± 0.27 g/(m2 d) due to the synergistic effect of ISs and plant biomass. Plant biomass could promote the ISs release efficiency, while ISs could facilitate plant biomass availability by promoting cellulose decomposition. High-throughput sequencing analysis revealed that the iron-oxidizing bacteria Pseudomonas were the dominant genus in FeB-EFB. Meanwhile, the existence of plant biomass could increase the abundance of iron-related bacteria and enrich heterotrophic and facultative denitrifying bacteria (e.g., Hydrogenophaga, Comamonas) as well, improving iron-mediated denitrification and heterotrophic denitrification simultaneously. Therefore, mixotrophic denitrification improvement played a major role in promoting nitrogen removal of FeB-EFB. These results indicated that coupling iron scraps with plant biomass may be an effective way to improve the nitrogen removal performance of EFB.

Microbial response mechanism of plants and zero valent iron in ecological floating bed: Synchronous nitrogen, phosphorus removal and greenhouse gas emission reduction.

Iron-based ecological floating beds (EFBs) are often used to treat the secondary effluent from wastewater treatment plant to enhance the denitrification process. However, the impact and necessity of plants on iron-based EFBs have not been systematically studied. In this research, two iron-based EFBs with and without plants (EFB-P and EFB) were performed to investigate the response of plants on nutrient removal, GHG emissions, microbial communities and functional genes. Results showed the total nitrogen and total phosphorus removal in EFB-P was 45-79% and 48-72%, respectively, while that in EFB was 31-67% and 44-57%. Meanwhile, plants could decrease CH4 emission flux (0-3.89 mg m-2 d-1) and improve CO2 absorption (4704-22321 mg m-2 d-1). Plants could increase the abundance of Nitrosospira to 1.6% which was a kind of nitrifying bacteria dominant in plant rhizosphere. Among all denitrification related genera, Simplicispira (13.08%) and Novosphingobium (6.25%) accounted for the highest proportion of plant rhizosphere and iron scrap, respectively. Anammox bacteria such as Candidatus_Brocadia was more enriched on iron scraps with the highest proportion was 1.21% in EFB-P, and 2.20% in EFB. Principal co-ordinates analysis showed that plants were the critical factor determining microbial community composition. TN removal pathways were mixotrophic denitrification and anammox in EFB-P while TP removal pathways were plant uptake and phosphorus-iron coprecipitation. In general, plants play an important directly or indirectly role in iron-based EFBs systems, which could not only improve nutrients removal, but also minimize the global warming potential and alleviate the greenhouse effect to a certain extent.

Simultaneous denitrification and iron-phosphorus precipitation driven by plant biomass coupled with iron scraps in subsurface flow constructed wetlands.

This study investigated the interaction between plant biomass and iron scraps and their influence on nitrogen (including nitrate and ammonia) and phosphorus removal in the subsurface flow constructed wetland. The results showed that with the addition of 0.5 g L-1 of plant biomass and 5.0 g L-1 of iron scraps, the nitrate, total nitrogen and total phosphorus removal were simultaneously improved. During 35 days of continuous operation, the plant biomass played main effect on the enhanced denitrification, accounting for about 57%, while iron scraps enhanced the other 43% of nitrogen removal and most phosphorus removal through precipitation inside the wetlands. Iron scraps could benefit the degradation of cellulose into low molecular carbohydrates by 10%, and the biomass could promote the oxidation of iron and increase the total phosphorus removal by 15%. Plant biomass coupled with iron scraps also improved simultaneously the richness, diversity and evenness of microbial community and promoted the abundance of Nitrospira (17.37%) and Thiobacillus (8.46%) in wetlands. In practice, putting iron scraps as matrix and placing plant biomass in the influent region would be a better choice. This research would provide a new method for effective utilization of plant biomass and iron scraps and further treatment of low-polluted wastewater in the wetlands.

The suitable biomass carbon source for improving nitrogen removal in surface flow constructed wetland system: Fresh vs. withered.

Plant biomass can be used as a carbon source to enhance the nitrogen removal effect. Related researches mainly focused on withered biomass, while little on fresh biomass. In this study, batch experiments revealed that the carbon release rate (60.5 mg TOC g-1) of fresh biomass was significantly higher than that of withered biomass (44.9 mg TOC g-1), while the nitrogen release rate showed the opposite trend. Compared with withered biomass, fresh biomass could release more sugar and less refractory humic acid, which means higher bioavailability. After adding fresh biomass, TN removal rate increased from 18.65% to 51.59%. The TN removal in the wetland adding withered biomass increased from 13.59% to 42.25%. The biomass addition had a slight impact on the effluent sensory quality in the first two days. After adding fresh biomass and withered biomass, the relative abundance of denitrifying bacteria like Flavobacterium and Pseudomonas in the system significantly increased. In general, fresh biomass was more suitable as a denitrification carbon source for surface flow constructed wetlands.

Efficient utilization of Iris pseudacorus biomass for nitrogen removal in constructed wetlands: Combining alkali treatment.

Aquatic plant biomass like Iris pseudacorus can be used as electron donor to improve denitrification performance in subsurface constructed wetlands. However, the phenomenon that the nitrogen removal rate declined in the terminal stage restricted the utilization of litters. In terms of this problem, this study investigated the performance of the used biomass through alkali treatment on nitrogen removal and analyzed the effect of alkali treatment on the component and structure of biomass and microbial community. The results showed that the alkali-treated biomass could further enhance the nitrogen removal by nearly 15% compared with used ones. The significant damage of cell walls and compact fibers containing cellulose and lignin through alkali treatment mainly resulted in the improvement of carbon release and nitrogen removal. With the addition of alkali-treated biomass, the richness index of microbes was higher compared with other biomass materials. Furthermore, the abundance of denitrification related genera increased and the abundance of genera for nitrification was maintained. Based on these finds, a mode of a more efficient Iris pseudacorus self-consumed subsurface flow constructed wetlands was designed. In this mode, the effluent total nitrogen could be stabilized below 5 mg L-1 for nine months and the weight of litters could be further cut down by 75%. These findings would contribute to efficient utilization of plant biomass for nitrogen removal enhancement and final residue reduction in the wetlands.

Zero-valent iron coupled plant biomass for enhancing the denitrification performance of ecological floating bed.

The ecological floating bed (EFB) coupled with zero-valent iron (ZVI) is proposed to treat low carbon-to-nitrogen ratio water. However, the application of ZVI is limited by low electron transfer efficiency. Coupling ZVI with carbon materials may improve the performance. In this study, the EFB with ZVI coupled plant biomass (IB-EFB) was established to enhance denitrification performance and compared to the EFB with ZVI coupled activated carbon (IC-EFB). The results showed that higher denitrification rate was observed in IB-EFB (68.8%) than that in IC-EFB (54.40%), which attributed to the synergistic effect of ZVI and plant biomass. Plant biomass also promoted the electron transfer of ZVI which enhanced the Fe(II)-mediated denitrification. High-throughput sequencing analysis revealed that IB-EFB enriched iron-related denitrifying bacteria more effectively than IC-EFB, and obtained high abundance of phototrophic Fe(II)-oxidizing bacteria Rhodopseudomonas (19.26%). Thus coupling ZVI with plant biomass has a potential for enhanced nitrogen removal in EFB.

Response mechanism of different electron donors for treating secondary effluent in ecological floating bed.

Electron donors have been widely used to improve denitrification performance. However, it is controversial which electron donor could be chosen. In this study, three electron donors were used to improve nitrogen removal from ecological floating beds (EFBs). The results showed that TN removal efficiency was 49-80%, 46-81%, and 45-79% in EFB-C (sodium acetate), EFB-S (sodium thiosulfate), EFB-Fe (iron scraps), respectively. Nitrification was limited in EFB-C and EFB-S while denitrification in EFB-Fe. The TN removal in the three EFBs were almost equivalent when HRT was 3 days. Lowest CH4 and N2O emissions were measured in EFB-Fe. Nitrifying and denitrifying bacteria were mainly concentrated in the root rhizospheres while iron cycle related and anammox bacteria were mainly concentrated on iron scraps surface. Heterotrophic denitrification and autotrophic denitrification were mainly attributed to TN removal in EFB-C and EFB-S, respectively. Autotrophic, heterotrophic denitrification and anammox contributed to TN removal in EFB-Fe.

Biological iron nitrogen cycle in ecological floating bed: Nitrogen removal improvement and nitrous oxide emission reduction.

Ecological floating beds (EFBs) have become a superior method for treating secondary effluent from wastewater treatment plant. However, insufficient electron donor limited its denitrification efficiency. Iron scraps from lathe cutting waste consist of more than 95% iron could be used as electron donors to enhance denitrification. In this study, EFBs with and without iron scraps supplementation (EFB-Fe and EFB, respectively) were conducted to explore the impacts of iron scraps addition on nitrogen removal, nitrous oxide (N2O) emissions and microbial communities. Results showed the total nitrogen (TN) removal in EFB-Fe improved to 79% while that in EFB was 56%. N2O emission was 0-6.20 mg m-2 d-1 (EFB-Fe) and 1.74-15.2 mg m-2 d-1 (EFB). Iron scraps could not only improve nitrogen removal efficiency, but also reduce N2O emissions. In addition, high-throughput sequencing analysis revealed that adding iron scraps could improve the sum of denitrification related genera, among which Novosphingobium accounted for the highest proportion (6.75% of PFe1, 4.24% of PFe2, 3.18% of PFe3). Iron-oxidizing bacteria and iron-respiring bacteria associated with and nitrate reducing bacteria mainly concentrated on the surface of iron scraps. Principal co-ordinates analysis (PCoA) indicated that iron scraps were the key factor affecting microbial community composition. The mechanism of iron scraps enhanced nitrogen removal was realized by enhanced biological denitrification process. Iron release dynamic from iron scraps was detected in bench-scale experiment and the electron transfer mechanism was that Fe0 transferred electrons directly to NO3--N, and biological iron nitrogen cycle occurred in EFB-Fe without secondary pollution.

Recycled utilization of Iris pseudacorus in constructed wetlands: Litters self-consumption and nitrogen removal improvement.

Aquatic plants litters from constructed wetlands might become pollutants without proper treatment. Due to its high carbon and low nitrogen contained, Iris pseudacorus litters have potential to be used as carbon source to enhance denitrification process in advanced treatments of secondary effluent from wastewater treatment plants. This study investigated the characteristics of carbon release form Iris pseudacorus litters and its performance on enhancement of nitrogen removal. The batch experiment showed that the organic carbon release process can be simulated by combining dissolution and hydrolysis process, and it was found that dissolved organic matters mainly consisted of 60% sugar and 35% humic acid-like compounds from the neutral detergent solution and hemicellulose of litters. The long-term operation of lab-scale constructed wetlands revealed a high nitrogen removal of 78.81-90.39% in treating the synthetic wastewater treatment plants effluent with the equivalent dosage of 25-150 g litters m-2 d-1. Furthermore, it is possible to establish an Iris pseudacorus self-consumed constructed wetland to reuse all of the litters produced during the operation. These findings can contribute to the understanding of the dynamics of carbon release from Iris pseudacorus litters and recycled utilization of plant biomass in the constructed wetlands.

Electrons transfer determined greenhouse gas emissions in enhanced nitrogen-removal constructed wetlands with different carbon sources and carbon-to-nitrogen ratios.

A constructed wetland (CW) was established to explore the influence of carbon addition (glucose or sodium acetate) on nitrogen removal and greenhouse gas (GHG) emissions at chemical oxygen demand to nitrogen ratios (COD/Ns) of 0, 4, 7. Results showed that the type of carbon source and COD/N significantly influenced the CW performance, in which the electrons transfer determined the regulation of denitrification, methanogenesis and respiration. Higher N2O emissions were consistent with higher nitrite accumulation at low COD/N because of electrons competition. The residual carbon source after near-complete denitrification could be further utilized by methanogenesis. Sodium acetate was superior to glucose in promoting denitrification and reducing global warming potential (GWP). In addition, bacteria sequencing and functional genes confirmed the important role of the type of carbon source on controlling nitrogen removal, carbon consumption and GHG emissions in microbial communities.

Denitrification- and anammox-dominant simultaneous nitrification, anammox and denitrification (SNAD) process in subsurface flow constructed wetlands.

Simultaneous nitrification, anammox and denitrification (SNAD) process was developed in subsurface flow constructed wetlands (SFCWs) to treat polluted surface water. The effects of vegetation, hydraulic retention time (HRT), C/N, and influent nitrogen forms on nitrogen removal and microbial communities were investigated. Results showed that denitrification- and anammox-dominant SNAD corresponded to nitrate- and ammonia-dominant influent, respectively, and both could achieve more efficient nitrogen removal in planted SFCWs than the unplated. These higher efficiencies were due to the microbial growth, organic carbon release, oxygen supply and plant uptake promoted by vegetation. The electron donors accelerated denitrification but inhibited ammonia oxidation with deficient oxygen. Anammox contributed to nitrogen removal of 27.34% under oxygen-limited conditions without vegetation. Anammox combined with denitrification and plant uptake were over 90% in planted SFCWs. For the investigated factors, the ammonia, nitrate and C/N were the most significant ones influencing the microbial communities, further nitrogen removal pathways and performances.