How do aeration mode and influent carbon/nitrogen ratio affect pollutant removal, gas emission, functional genes and bacterial community in subsurface wastewater infiltration systems?

The discharge of decentralized domestic wastewater has resulted in the contamination of surface water and groundwater, causing ecological damage and exacerbating water scarcity in numerous countries (Yang et al. 2016). However, traditional centralized wastewater treatment technologies are unfit for decentralized domestic wastewater due to high construction, operation investment and complex maintenance (Ji et al. 2012). Subsurface wastewater infiltration system (SWIS) is a kind of low-construction cost and easy management ecological treatment method, which has been widely applied to treat decentralized domestic wastewater in many areas (Zheng et al. 2018). The construction cost of SWIS is one-half of the secondary biochemical treatment and the operating cost is one-fifth (Li et al. 2013). In subsurface wastewater infiltration systems (SWISs), wastewater is purified by physical, chemical and biochemical action after entering into the underground soil layer. The removal efficiency of COD and TP was more than 80 and 90%, respectively (Li et al. 2021). However, TN removal efficiency was only around 50% which limited its application (Li et al. 2020).

Biological nitrification and denitrification is the major nitrogen removal process in SWISs. The strategies widely adopted to strengthen nitrogen removal are: (1) amending the original soil with different materials, such as dewatered sludge, biochar and functional microorganisms (Sun et al. 2018; Jia et al. 2019; Liang et al. 2019); (2) increasing carbon source supply for denitrification process (Li et al. 2021); (3) improving soil oxygen environment for nitrification by aeration or intermittent operation (Pan et al. 2015; Chen et al. 2021b). Among these methods, aeration has been acknowledged as an effective and practical approach for enhancing the oxygen environment within the matrix and facilitating nitrogen removal (Fan et al. 2013; Liang et al. 2019). Aeration mode could be categorized into continuous aeration and intermittent aeration (IA), which could create different matrix oxygen environments and ultimately affect nitrogen and organics removal (Yang et al. 2016). The influent carbon/nitrogen ratio affected nitrogen removal in SWISs significantly (Li et al. 2018). Relevant studies have predominantly focused on the individual effects of aeration or carbon/nitrogen ratio on the removal of pollutants in SWISs (Zheng et al. 2018; Li et al. 2020).

The emissions of CO₂, CH₄ and N₂O from SWISs are significant contributors to the greenhouse effect and have garnered increasing attention in recent years (Kong et al. 2016). CO₂ is emitted from the aerobic degradation of organics. CH₄ occurs in the anaerobic fermentation of organics and N₂O is produced in the process of biological nitrification and denitrification. Many factors could affect greenhouse gas emissions in SWISs. Li et al. (2019) and Pang et al. (2020) found that influent hydraulic loading, pollutant loading and concentration affected N₂O release in SWISs. CH₄ and N₂O emissions increased with the increase of influent carbon/nitrogen ratio, matrix oxidation reduction potential (ORP) and temperature in non-aeration SWISs, which was reported by Kong et al. (2016). Sun et al. (2018) and Zheng et al. (2018) concluded IA was a feasible strategy for SWISs to reduce N₂O generation. However, there are few researches absorbed in the influence of influent carbon/nitrogen ratio and aeration mode on CO₂, CH₄ and N₂O emission in SWISs synchronously.

In this study, series of laboratory-scale experiments in SWISs was carried out to (1) reveal the influences of influent carbon/nitrogen ratio and aeration mode on organic removal, nitrogen removal and CO₂, CH₄ and N₂O emission; (2) explore bacterial community, vertical distribution and the abundances of functional genes involved in COD removal, removal, TN removal, three greenhouse gases emission with different influent carbon/nitrogen ratios and aeration modes; (3) achieve optimal operation parameters with which best pollutants removal were achieved and least greenhouse gases were produced.

Synthetic wastewater was continuously fed for each SWIS with 24 h hydraulic retention time. The carbon/nitrogen ratio of synthetic wastewater was adjusted by altering the content of sodium acetate (CH₃COONa). Four different carbon/nitrogen ratios (a: 4; b: 8; c: 12; d: 16) were formulated. Synthetic wastewater also composed of 165 mg/L (NH₄)₂SO₄; 15 mg/L KH₂PO₄; 8 mg/L NaNO₃; 12 mg/L MgSO₄; 12 mg/L MnSO₄; 12 mg/L ZnSO₄·7H₂O; 12 mg/L FeSO₄ and 12 mg/L CaCl₂. The characteristics of the synthetic wastewater was shown in Table 1.

Table 1

The characteristics of the synthetic wastewater (mean ± SD)

Carbon/nitrogen ratio .	4 .	8 .	12 .	16 .
COD (mg/L)	163.6 ± 8.9	324.4 ± 14.5	487.1 ± 10.7	658.9 ± 18.2
(mg/L)	38.5 ± 0.6	39.2 ± 0.3	38.1 ± 0.5	38.7 ± 0.2
TN (mg/L)	41.3 ± 0.3	41.7 ± 0.5	41.8 ± 0.6	41.1 ± 0.7
(mg/L)	1.3 ± 0.6	1.2 ± 0.3	1.5 ± 0.2	1.1 ± 0.4
TP (mg/L)	4.2 ± 0.4	4.1 ± 0.2	4.5 ± 0.3	4.4 ± 0.1

Carbon/nitrogen ratio .	4 .	8 .	12 .	16 .
COD (mg/L)	163.6 ± 8.9	324.4 ± 14.5	487.1 ± 10.7	658.9 ± 18.2
(mg/L)	38.5 ± 0.6	39.2 ± 0.3	38.1 ± 0.5	38.7 ± 0.2
TN (mg/L)	41.3 ± 0.3	41.7 ± 0.5	41.8 ± 0.6	41.1 ± 0.7
(mg/L)	1.3 ± 0.6	1.2 ± 0.3	1.5 ± 0.2	1.1 ± 0.4
TP (mg/L)	4.2 ± 0.4	4.1 ± 0.2	4.5 ± 0.3	4.4 ± 0.1

View Large

Three laboratory-scale SWISs were named NA-SWIS, IA-SWIS and CA-SWIS, respectively. IA-SWIS and CA-SWIS contained aeration units which consisted of air pump, gas meter and aeration diffuser. The aeration diffuser was at 0.6 m below the surface matrix. Aeration diffusers and influent distributing pipes were wrapped in gravel with a diameter of 1–2 cm to prevent clogging. The airflow rate of 3.5 ± 0.2 L/min was applied in IA-SWIS and CA-SWIS, respectively. The cyclic aeration/non-aeration process was for IA-SWIS, which was 1 h for aeration and then 5 h without aeration. Continuous aeration was for CA-SWIS and non-aeration was for NA-SWIS. Before this experiment, three SWISs have been run for 45 days to allow microorganisms maturation. Four running plans were arranged in each SWIS with a carbon/nitrogen ratio gradually elevated from 4 to 8, 12 and 16. An increase in the influent carbon/nitrogen ratio also meant an increase in the organic loading rate. Each running plan lasted for 40 days. The experiment commenced in May of 2022 and spanned over a duration exceeding 5 months.

Water samples and gas samples were collected after each SWIS had been run for 10 days under the experimental carbon/nitrogen ratio. The mature SWIS takes about a week to operate steadily under a new running parameter (Li et al. 2020).

Water samples were taken from the influent and outlet every 5 days. COD, ⁠, ⁠, ⁠, TOC and TN of the water samples were analyzed on the basis of water and wastewater standard analytical methods of China (Standard Method for the Examination of Water & Wastewater Editorial Board 2002).

ρ represents CO₂, CH₄ or N₂O mass concentration, mg/m³; φ_v is the gas sample volume fraction; θ is the gas sample temperature, °C; M represents CO₂, CH₄ or N₂O molar mass, g/mol; H is the height of gas collection unit, m; Δt is the sampling interval time; ρ₂–ρ₁ represents CO₂, CH₄ or N₂O mass concentrations in sampling interval, mg/m³; m₁ represents CO₂, CH₄ or N₂O emission in sampling interval, mg; m₂ is the mass of TOC or TN removed in sampling interval, mg.

After each carbon/nitrogen ratio experiment, matrix samples were collected from sampling ports to analyze bacteria and functional genes by real-time quantitative polymerase chain reaction (qPCR) on 16S rRNA fragment of bacteria, the target fragment of ammonia monooxygenase (amoA) and nitrite oxidoreductase (nxrA), membrane-bound nitrate reductase (narG), periplasmic nitrate reductase (napA), nitrite reductase (nirS and nirK), nitric oxide reductase (norB), nitrous oxide reductase (nosZ) and Methyl Coenzyme M Reductase A (mcrA) genes which related to organics removal, nitrogen removal and CO₂, N₂O or CH₄ production. The specific procedures of qPCR were referred to the study of Ji et al. (2012). In addition, the bacterial community of each sample was analyzed by high-throughput sequencing technology according to Chen et al. (2021a).

Matrix oxygen concentrations were recorded and stored in data collectors every 20 min. The average value of matrix oxygen concentrations was calculated as the mean of all data collected after each SWIS was run for 10 days under the experimental carbon/nitrogen ratio.

Each sample was analyzed three times and the mean value of the three analyses was reported. Statistical analysis was performed using SPSS 19.0 software. The significance level was set at P < 0.05 and differences were assessed using two-way ANOVA.

In NA-SWIS, average removal efficiencies of were 84.7, 80.1, 47.7 and 15.6% with average effluent concentrations of 5.9, 7.6, 20.1 and 32.5 mg/L under carbon/nitrogen ratio of 4, 8, 12 and 16, respectively. Average effluent concentrations with carbon/nitrogen ratios of 12 and 16 exceeded the Class I of Municipal Sewage Treatment Plant Discharge Standard of China (⁠ ≤ 15 mg/L) (GB18918-2002). removal efficiencies of NA-SWIS decreased with the carbon/nitrogen ratio increasing. Nitrification is an aerobic reaction, which is the first process for removal (Song et al. 2016). Non-aeration SWIS could not achieve this reaction efficiently due to a lack of oxygen (Yang et al. 2016; Li et al. 2021). More oxygen was required with carbon/nitrogen ratio increasing which inhibited ammonia oxidation microorganisms from oxidizing by oxygen competition. Average removal efficiencies of in IA-SWIS and CA-SWIS were significantly higher than those in NA-SWIS with the same carbon/nitrogen ratio (P < 0.05). Average removal efficiencies decreased when the carbon/nitrogen ratio increased from 8 to 16 in IA-SWIS and CA-SWIS. Aeration could not supply enough oxygen for nitrification with a high carbon/nitrogen ratio in this study. However, the average removal efficiency of in CA-SWIS was significantly higher than that in IA-SWIS with a carbon/nitrogen ratio of 16 (P < 0.05) because more oxygen was provided by continuous aeration.

In SWISs, denitrification is widely regarded as the main way to nitrogen removal, which needs anaerobic conditions and enough carbon source (Li et al. 2021). TN removal in SWISs with aeration and without aeration was significantly different. Average TN removal efficiencies decreased with the carbon/nitrogen ratio increasing in NA-SWIS, which was the same trend with removal. With a high carbon/nitrogen ratio, removal was restrained, which limited the denitrification reaction due to inadequate supply as electron acceptors. In NA-SWIS, average removal efficiencies of TN were 44.6 and 16.4% and average effluent TN concentrations were 22.7 and 34.6 mg/L with carbon/nitrogen ratio of 12 and 16, which exceeded the above-mentioned standard (TN ≤ 20 mg/L). The carbon/nitrogen ratio of 4 and 8 was recommended for non-aeration SWIS in this study. With carbon/nitrogen ratio of 4 and 8, average removal efficiencies of TN in IA-SWIS and CA-SWIS were significantly lower than those in NA-SWIS (P < 0.05), which were 23.3 and 53.1% for IA-SWIS and 22.8 and 52.6% for CA-SWIS, respectively. Moreover, was the main component of TN in the effluents in IA-SWIS and CA-SWIS. After effective organics oxidation and nitrification by aeration, the carbon source as electron donor was deficient, which led as electron accepters to be accumulated and not to be eliminated with a low carbon/nitrogen ratio. The results accorded with the report of Fan et al. (2013). With a carbon/nitrogen ratio of 12 and 16, the average removal efficiencies of TN in IA-SWIS and CA-SWIS were significantly higher than those in NA-SWIS (P < 0.05). After effective nitrification in aeration SWISs, a high carbon/nitrogen ratio provided more carbon sources for denitrification and resulted in less accumulation, which achieved high TN removal. The average TN removal efficiency of IA-SWIS was significantly higher than that of CA-SWIS with a carbon/nitrogen ratio of 12 due to more carbon sources obtained in IA-SWIS (P < 0.05). However, with a carbon/nitrogen ratio of 16, the average removal efficiency of TN in IA-SWIS was significantly lower than that in CA-SWIS because lower nitrification of IA-SWIS achieved compared with CA-SWIS and more carbon source provided after high efficient nitrification in CA-SWIS (P < 0.05). A carbon/nitrogen ratio of 12 and 16 was suggested for IA-SWIS and CA-SWIS separately.

CH₄ conversion efficiencies of NA-SWIS enhanced with carbon/nitrogen ratio increasing, which were 0.82, 0.94, 1.12 and 1.35% with average CH₄ emission rates of 9.4, 21.4, 37.5 and 58.2 mg/(m²·day) under carbon/nitrogen ratio of 4, 8, 12 and 16, respectively (in Figure 4). The increased carbon/nitrogen ratio meant more organic matter, which could expend more oxygen and supply more carbon source for the growth of anaerobic methanogen in non-aeration systems (Li et al. 2018). As a result, CH₄ conversion efficiencies and emission rates increased with the influent carbon/nitrogen ratio increasing in NA-SWIS. Average CH₄ conversion efficiencies were below 0.68% with carbon/nitrogen ratios of 4, 8, 12 and 16 in IA-SWIS and CA-SWIS, which were significantly lower than those in NA-SWIS with the same carbon/nitrogen ratio (P < 0.05). Low emission rates of CH₄ in IA-SWIS and CA-SWIS were attributed to aeration, which improved organic aerobic oxidation and led to a decrease of available organics to methanogen. Wang et al. and Yang et al. reported the same result that aeration could reduce the abundance of methanogen and decrease CH₄ production. Average CH₄ conversion efficiencies of IA-SWIS and CA-SWIS were not significantly different (Yang et al. 2017). In aeration SWISs, there were no significant differences in CH₄ conversion efficiencies among different carbon/nitrogen ratios in this study. It could be concluded that aeration mode and carbon/nitrogen ratio could hardly affect CH₄ emission in aeration SWISs.

Average N₂O emission rates were 5.2, 4.8, 3.2 and 2.4 mg/(m² day) with conversion efficiencies of 0.89, 0.78, 0.55 and 0.41% when the carbon/nitrogen ratio was 4, 8, 12 and 16 in NA-SWIS, respectively (in Figure 4). Furthermore, average N₂O conversion efficiencies decreased with the carbon/nitrogen ratio increasing, which was in agreement with the report of Li et al. (2018). A higher carbon/nitrogen ratio represents more organics, which might consume more available oxygen to oxide it and further constrain the autotrophic ammonia oxidation bacteria to oxidize due to oxygen competition. The final result was that denitrification as the next step of nitrification was also restricted in NA-SWIS. Therefore, N₂O emission decreased. Jacobs and Harrison concluded when carbon source availability was high, nitrifying bacteria were poor competitors for compared with heterotrophic microbes (2014). Sabba et al. (2015) and Maucieri et al. (2017) also revealed that N₂O conversion efficiency positively correlated with nitrification rate in low denitrification rate wastewater treatment systems. Average N₂O emission rates and conversion efficiencies of CA-SWIS and IA-SWIS were more than 5.7 mg/(m² day) and 0.97% with carbon/nitrogen ratios of 4 and 8, which were significantly higher than those of NA-SWIS because of high nitrification rates and low denitrification rates in aeration SWISs (P < 0.05). After efficient nitrification and aerobic oxidation of organics, the carbon source was not enough for the denitrification process which constrained N₂O transforming to N₂ in IA-SWIS and CA-SWIS with low carbon/nitrogen ratios of 4 and 8. Average N₂O emission rates and conversion efficiencies with a carbon/nitrogen ratio of 12 and 16 were significantly lower than those with a carbon/nitrogen ratio of 4 and 8 in aeration SWISs (P < 0.05). Similar findings were reported by Li et al. (2018), Wang et al. (2014) and Zhou et al. (2017). A higher carbon/nitrogen ratio provided more organics for N₂O to N₂ transformation, which led to N₂O emission reduction. The average emission rate of N₂O in IA-SWIS was lower than that in CA-SWIS with a carbon/nitrogen ratio of 12 and was higher than that in CA-SWIS with a carbon/nitrogen ratio of 16. Compared with CA-SWIS, IA-SWIS could obtain more carbon sources with a carbon/nitrogen ratio of 12, which improved N₂O transforming to N₂. The concentrations of and TN in the effluent were 0.8 and 6.9 mg/L for IA-SWIS, which exhibited significantly lower than that in CA-SWIS (P < 0.05). The result indicated that denitrification converted more to N₂ in IA-SWIS with a carbon/nitrogen ratio of 12. The effluent concentrations of and TN were 1.3 and 7.2 mg/L for CA-SWIS, which were significantly lower than those of IA-SWIS (P < 0.05) with a carbon/nitrogen ratio of 16. More carbon sources were provided after higher efficient nitrification in CA-SWIS with a carbon/nitrogen ratio of 16 in comparison with IA-SWIS, which reduced N₂O emission and converted more to N₂ in CA-SWIS.

A carbon/nitrogen ratio of 12 and 16 were separately proposed for IA-SWIS and CA-SWIS in consideration of CH₄ and N₂O emission.

In NA-SWIS, IA-SWIS and CA-SWIS, the bacterial 16S rRNA abundances decreased with matrix depth increasing, which complied with the same trend of matrix oxygen concentration. In NA-SWIS, the bacterial 16S rRNA abundances at 0.6 m depth decreased with the carbon/nitrogen ratio increasing because of insufficient oxygen supply. In contrast, the bacterial 16S rRNA abundances at 0.6 m depth increased with the carbon/nitrogen ratio increasing in IA-SWIS and CA-SWIS. With the same carbon/nitrogen ratio, the bacterial 16S rRNA abundances in IA-SWIS and CA-SWIS were significantly higher than those in NA-SWIS at 0.6 m depth (P < 0.05), which could interpret higher organics removal and more CO₂ emission in aeration SWISs. Aeration supplied excess oxygen and a high carbon/nitrogen ratio provided more organic substrate for bacteria which promoted bacterial growth and reproduction (Pan et al. 2015). The bacterial 16S rRNA abundances in CA-SWIS were higher than those in IA-SWIS at 0.6 m depth because of more oxygen supplied by continuous aeration with carbon/nitrogen ratio of 12 and 16.

In aeration and non-aeration SWISs, the mcrA abundances increased with matrix depth increasing, which was opposite the tendency of matrix oxygen. The mcrA abundances at 0.6, 0.9 and 1.2 m depths increased with carbon/nitrogen ratio increasing in NA-SWIS. With carbon/nitrogen ratio increasing, methanogen could acquire more organics and the oxygen supply became insufficient, which helped to methanogen. The mcrA abundances in NA-SWIS were significantly higher than those in IA-SWIS and CA-SWIS with the same depth and carbon/nitrogen ratio (P < 0.05), which accorded with CH₄ emission. CH₄ emission rate was positively correlated with the mcrA abundance (Morris et al. 2014). There were no significant differences between the mcrA abundances in CA-SWIS and IA-SWIS with the same depth and carbon/nitrogen ratio in this study (P > 0.05).

In NA-SWIS, IA-SWIS and CA-SWIS, the amoA and nxrA abundances decreased with matrix oxygen decreasing. In NA-SWIS, the amoA and nxrA abundances at 0.6 m depth decreased with carbon/nitrogen ratio increasing because of insufficient oxygen supply, which was in accordance with removal. In IA-SWIS and CA-SWIS, the amoA and nxrA abundances were not significantly different with carbon/nitrogen ratio of 4 and 8 due to enough oxygen supply (P > 0.05). The amoA and nxrA abundances were hardly affected by aeration mode. However, the amoA and nxrA abundances decreased when carbon/nitrogen ratio increased from 8 to 12 and 16. High available organics restricted the nitrification process in SWISs with/without aeration (Song et al. 2016; Li et al. 2018). The amoA and nxrA abundances at 0.6 m depth of IA-SWIS and CA-SWIS were significantly higher than those of NA-SWIS (P < 0.05) due to aeration, which followed removal.

In NA-SWIS, the narG, napA, nirK, nirS, norB and nosZ abundances at 0.9 and 1.2 m depths decreased with carbon/nitrogen ratio increasing. Nitrification reaction was constrained with carbon/nitrogen ratio increasing, which led to insufficient for denitrification and decreased the enrichment of six functional genes of the denitrification process. When carbon/nitrogen ratios were 4 and 8, the abundances of six functional genes in NA-SWIS at 0.9 and 1.2 m depths were higher than those in IA-SWIS and CA-SWIS. Carbon source insufficient with low carbon/nitrogen ratio in aeration SWISs decreased the abundances of six functional genes. In aeration, SWISs, the abundances of six functional genes were significantly higher than those in NA-SWIS at 0.9 and 1.2 m depths with carbon/nitrogen ratio of 12 and 16 (P < 0.05), which could clarify excellent TN removal and N₂O emission in IA-SWIS and CA-SWIS with high carbon/nitrogen ratio. In this study, aeration improved nitrification efficiency and a high carbon/nitrogen ratio provided more carbon source for denitrification, which enhanced the narG, napA, nirK, nirS, norB and nosZ abundances with carbon/nitrogen ratios of 12 and 16.

Table 2 shows the alpha diversity of different depth matrices in three SWISs. The Chao1 and Shannon indexes reflected the bacterial species richness and bacterial community diversity, respectively (Chen et al. 2021b). The higher Chao1 and Shannon values indicated higher species richness and community diversity in matrix samples. The Chao1 and Shannon values of CA-SWIS and IA-SWIS were significantly higher than those of NA-SWIS, which indicated that higher bacterial species richness and community diversity were achieved in CA-SWIS and IA-SWIS. The facilitated proliferation of bacteria in CA-SWIS and IA-SWIS was attributed to aeration. The richness and diversity of bacteria decreased with the depth increasing in three SWISs. Yang et al. (2021) drew the same conclusion that more bacterial species and higher bacterial community diversity were found in aerobic conditions than anoxic/anaerobic conditions in SWISs.

The aeration treatment effectively developed an aerobic environment in the matrix above 0.6 m, while maintaining an anoxic/anaerobic environment below 0.9 m. The aeration mode and carbon/nitrogen ratio exerted significant influences on the removal of organic compounds, nitrogen, CO₂, CH₄ and N₂O emissions, as well as the composition of bacterial communities. COD, ⁠, TN removal efficiencies, CO₂ and N₂O conversion efficiencies decreased and CH₄ emission increased with carbon/nitrogen ratio increasing in non-aeration SWIS. With high carbon/nitrogen ratio of 12 and 16, aeration improved COD, ⁠, TN removal, the bacterial 16S rRNA, amoA, nxrA, narG, napA, nirK, nirS, norB, nosZ abundances, bacterial community Alpha diversity, the relative abundances of Actinobacteria at 0.6 m depth, the relative abundances of Chloroflexi, Gemmatimonadetes, Bacteroidetes and Firmicutes at 0.9 and 1.2 m depths and decreased CH₄ and N₂O conversion efficiencies, the mcrA abundances in IA-SWIS and continuous aeration SWIS compared with those in non-aeration SWIS. In consideration of the removal efficiencies for organics, nitrogen and CH₄ and N₂O conversion, carbon/nitrogen ratios of 4 and 8 are recommended for non-aeration SWIS. For IA-SWIS and continuous aeration SWIS, carbon/nitrogen ratios of 12 and 16 are proposed, respectively.

We gratefully thank the financial support from the Applied Basic Research Plan of Liaoning (2023JH2/101300053); Science and Technology Plan of Shenyang (21-108-9-36), Basic Research Plan of Liaoning Education Department (2023), Major Original Program in Shenyang Normal University (ZD201904), and The Ninth Batch of Education and Teaching Reform Project of Shenyang Normal University (JG2021-YB099).

L. Z. investigated the work, wrote the article and prepared the original draft. Z. C. developed the methodology, investigated the article and rendered support in data curation. B. Y. investigated the work and revised the article. Z. L. investigated the article and rendered support in data curation. J. P. wrote the review and edited the article, rendered support in funding acquisition and visualized the article.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.