Geographical pattern of methanogenesis in paddy and wetland soils across eastern China

Science of the Total Environment 651 (2019) 281–290

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Geographical pattern of methanogenesis in paddy and wetland soils across eastern China Xin Hao, Shuo Jiao, Yahai Lu ⁎ College of Urban and Environmental Sciences, Peking University, Beijing, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Lag phase for soil methanogenesis increases from north to south in China. • Depletion time of Fe(III) and sulfate positively correlated with methanogenesis lag phase. • Geographical variations in methanogenesis are related to the soil pH variation. • Greater temperature response for soil methanogenesis from south to north in China

a r t i c l e

i n f o

Article history: Received 9 August 2018 Received in revised form 8 September 2018 Accepted 13 September 2018 Available online 13 September 2018 Editor: Jay Gan Keywords: Methanogenesis Fermentation Paddy fields Wetlands Biogeography

a b s t r a c t Large variation of CH4 emissions from paddy and wetland ecosystems exists across different geographical locations in China. To obtain mechanistic understanding of this variation, we investigated the dynamics of methanogenesis over the course of glucose degradation in fourteen paddy field soils and five wetland soils collected from different regions of China. The results revealed that the maximal rate (2–3 mM per day) and the total amount (25–30 mM) of CH4 produced were similar across soil samples. The lag phase of methanogenesis, however, differed substantially with the shortest lag phase of 4 days in a paddy soil from north China and the longest of 32 days in a soil from south China, and this difference reflected a general geographical trend among all soils tested. Nitrate was reduced completely within 4 days in all soils. The reduction of Fe(III) and sulfate was completed after 21 days and 29 days, respectively. The depletion time of Fe(III) and sulfate were positively correlated with the lag phase of methanogenesis. Competition for common substrates between methanogens and iron and sulfate reducers, however, does not explain this coincidence because a slow production of CH4 was detected at the very beginning. It appears that the geographical variations in methanogenesis and the reduction of ferric iron and sulfate are related to the variation in soil pH but not to temperature, soil organic C and nutrient conditions in paddy and wetland soils across eastern China. © 2018 Elsevier B.V. All rights reserved.

⁎ Corresponding author at: Peking University, College of Urban and Environmental Sciences, No. 5, Yiheyuan Road, Haidian District, Beijing 100871, China. E-mail address: [email protected] (Y. Lu).

https://doi.org/10.1016/j.scitotenv.2018.09.167 0048-9697/© 2018 Elsevier B.V. All rights reserved.

282

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

1. Introduction Methane is the second important greenhouse gas next to CO2, with its atmospheric concentration approaching 1850 ppb in 2015, almost tripled since the preindustrial times (Saunois et al., 2016; Schaefer et al., 2016). Sources for global CH4 emissions include natural wetlands, termites and gas hydrates and anthropogenic paddy fields, landfills, livestock, biomass burning and fossil fuel production (Menon et al., 2007). Rice cultivation has been practiced in Asian countries for a long history and presently accounts for over 90% of annual world rice production. As a result, the Asian paddy fields represent the most important contributor to the global methane emissions associated with rice production (Bhatia et al., 2013; Chen et al., 2013; Zhang et al., 2016). Not only with a long history but rice cultivation is developed in Asia wherever the water availability is proper. Paddy fields are thus widespread geographically from low to high latitudes and from low to high altitudes. Albeit similar cultivation practice across different regions there exist large variations in landscapes, soil properties, climate conditions, farmer's managements, and cultivation histories. Consequently, large geographical variations in processes and functions of paddy ecosystems are expected. Among these processes, methanogenesis is of global change significance, and yet very little has been known. Understanding the geographical variation of CH4 emissions and deciphering the underlined mechanisms shall help to shape a better knowledge for predicting and mitigating the CH4 emissions. Elucidating the biogeographical patterns of microbial communities has been a thematic focus in microbial ecology in recent years (Thompson et al., 2017). It has been recognized that the microbial community assembly is generally controlled by two types of ecological processes, i.e. stochastic and deterministic. How a particular community is formed in a given habitat depends on the balance of these processes. Dissecting the rates of various processes and estimating their contributions, however, remain as a hard task (Tripathi et al., 2018). Furthermore, many of previous studies on geographical distributions have been focused on total microbiomes with little attention paid on functional guilds (Nelson et al., 2016; Thompson et al., 2017). In fact, the elemental biogeochemical cycles are gauged mainly by the functional groups. To this end, though the large geographical difference in CH4 emissions from paddy fields has been demonstrated (Yan et al., 2003), only a few studies have evaluated the biogeographical distribution of methanogenic archaea in various environments. These studies showed that methanogenic community composition was significantly related to geographic distance, salinity, pH, temperature, and nutrients (Zu et al., 2016; Wen et al., 2017; Zhang et al., 2018). Our preliminary study, however, revealed weak distance-decay relationship of methanogenic archaea in Chinese paddy field and wetland soils (Zhang et al., 2018). Moreover, the translation of methanogenic community into CH4 production processes has yet to be established. Therefore, it is necessary to determine CH4 production potentials of soils from different locations in order to predict the geographical variation of CH4 emissions from paddy fields. In the present experiment, we collected fourteen paddy field soils across eastern China spanning a gradient of latitude from 18.6°N in south to 49.9°N in north. To increase the diversity of soil types for the evaluation of geographical pattern, five natural wetland soils were also collected. The paddy fields have intensively undergone anthropogenic activities, such as ploughing, fertilization and cyclic irrigation and drainage, whereas wetland is natural ecosystem with fewer disturbances of human being. The inclusion of wetland soils shall increase the robustness of geographical pattern if exists. CH4 production from anaerobic decomposition of organic matter is driven by a series of complicated microbial activities (Bridgham et al., 2013). Consequently, the pattern of methanogenesis is not only determined by the activity of methanogenic populations but also by the upstream fermentation (Drake et al., 2009) and the activities of anaerobically respiring microbes using electron acceptors other than CO2. In the latter case, competition for common substrates H2 and

acetate has been considered to occur between methanogens and the respiring anaerobes. This concept was first proposed by Winfrey and Zeikus (1977) and later validated in multifarious anaerobic environments (King, 1984; Lovley and Goodwin, 1988). In the light of thermodynamic theory, the electron acceptors are reduced sequentially in the order of nitrate, Mn(IV), Fe(III), sulfate and CO2 (Zehnder, 1988; Bridgham et al., 2013; Jelen et al., 2016). Consequently, methanogens are considered to be energetically most restricted and get activated only when those electron acceptors have been depleted (Lovley and Phillips, 1987; Thauer et al., 2008). It has been demonstrated that the addition of nitrate, sulfate and Fe(III) to active methane-producing soils suppressed methanogenesis (Achtnich et al., 1995a; Qu et al., 2004; Hori et al., 2009). However, if electron donor substrates are provided sufficiently, the different reduction processes may overlap and take place concomitantly (Achtnich et al., 1995a; Peters and Conrad, 1996; Megonigal et al., 2003). We hypothesized that large geographical distance may imply large variations of not only methanogenesis but also the processes of fermentation and reductions of oxidants other than CO2. To obtain a general understanding of methanogenesis, it is important to simultaneously analyze these processes. Therefore, the objectives of the present study were: i) to determine the dynamics of CH4 production and the reduction of oxidative ions (sulfate, nitrate and ferric iron) in nineteen soils collected across eastern China; and ii) to identify the important factors influencing the variation of methane production across different soils. During the experiment, soil samples were anaerobically incubated under identical laboratory conditions and glucose was added as organic substrate. 2. Materials and methods 2.1. Soil sampling Fourteen rice paddy soils and five wetlands samples were collected during the time from July to September in 2016 from different regions of China, with sample name abbreviation and geographic information listed in Table 1. At each site, five soil cores (0–20 cm in depth) with a distance of at least 5 m away from each other were collected, mixed thoroughly, and placed in ziplock plastic bags. Soil samples were shipped to lab within 24 h and stored at 4 °C. The chemical properties of the soils, including pH, total organic carbon (TOC), total nitrogen (TN), total phosphorus (TP), inorganic nitrogen (NO3−-N and NH4+-N), available phosphorus (AP) and available potassium (AK) were measured using the standard soil testing procedures (Bao, 2000). 2.2. Soil slurries preparation and incubation For anaerobic incubation, soil samples were taken out from refrigerator and allowed for 24 h recovery under room temperature. The soil samples were then suspended with autoclaved degassed water at a soil-water-ratio of 1:5 (1 g dry weight soil plus 5 ml autoclaved degassed water). Aliquots (20 ml) of homogenized soil slurry were transferred into 60 ml sterile serum bottles. Glucose (10 mM) was added into the bottles to serve as organic substrate. Bottles were evacuated and flushed with N2 at constant pressure for about 5 min and then capped with black butyl rubber stoppers and aluminum crimps. The bottles were vigorously shaken by hand to homogenize the soil slurry and then incubated under the dark at 30 °C without shaking. The incubation was set up with 40 to 60 replicates. At the predetermined time intervals, four replicates were sacrificed for the destructive sampling of soil slurries for chemical analyses as described below. 2.3. Chemical analyses Methane was analyzed by Agilent 7890B gas chromatography using a flame ionization detector (Zhang and Lu, 2016). Gas samples (0.1 ml)

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

283

Table 1 Site information, soil chemical properties and the maximal rate (Vmax) of CH4 production during anaerobic incubation. The maximal rate (Vmax) of CH4 production in each soil was calculated when the CH4 production reached to the linear increasing during the anaerobic incubation. Site

Site information

Soil chemical properties

Latitude Longitude MAT (°C)

Initial pH

HH (Heihe)

49.90°N

127.36°E

XKW (XingKai Lake) WC (Wuchang)

45.37°N

132.32°E

45.05°N

127.06°E

SY (Shenyang)

41.65°N

123.32°E

TJ (Tianjin)

39.83°N

117.35°E

HBW (Haibei Alpine) JN (Jining)

37.61°N

101.32°E

35.32°N

116.62°E

QD (Qidong)

31.88°N

121.72°E

THW (Tai Lake)

31.48°N

120.26°E

CD (Chengdu)

30.58°N

103.87°E

HZ (Hangzhou)

30.08°N

119.94°E

DTW (Dongting Lake) PYW (Poyang Lake) YY (Yiyang)

29.39°N

113.11°E

29.20°N

116.00°E

28.66°N

112.46°E

AS (Anshun)

26.05°N

105.77°E

GZ (Ganzhou)

25.98°N

116.11°E

DL (Dali)

25.29°N

100.24°E

ZS (Zhongshan)

22.68°N

113.44°E

LS (Lingshui)

18.62°N

109.96°E

0.2

5.97 ± 0.0 4.2 5.73 ± 0.0 4.6 6.76 ± 0.0 8.5 5.84 ± 0.0 12.6 8.01 ± 0.0 −1.2 7.66 ± 0.0 14.4 7.10 ± 0.0 15.5 6.96 ± 0.0 16.2 8.00 ± 0.0 16.3 7.21 ± 0.0 16.7 5.38 ± 0.0 17.5 5.46 ± 0.0 17.5 5.18 ± 0.0 17.4 4.42 ± 0.0 15.3 5.43 ± 0.0 19.2 5.29 ± 0.0 15.1 6.05 ± 0.0 22 5.41 ± 0.0 25 6.15 ± 0.0

Vmax (mM/d)

TOC (g/kg)

TN (g/kg)

TP (g/kg)

AN (mg/kg)

AP (mg/kg)

AK (mg/kg)

Initial NO− 3 (μΜ)

Initial SO2− 4 (μΜ)

Reducible Fe (III) (μmol g−1)

38.0 ± 6.9 241 ± 5.6 43.6 ± 1.2 23.6 ± 0.6 17.8 ± 1.0 211 ± 1.1 37.9 ± 1.4 31.8 ± 0.5 10.8 ± 0.7 38.5 ± 1.0 40.9 ± 1.0 22.3 ± 2.0 15.3 ± 0.8 40.0 ± 1.1 65.1 ± 1.0 18.2 ± 0.5 24.9 ± 0.5 25.2 ± 0.9 28.8 ± 0.6

1.67 ± 0.0 11.0 ± 0.1 1.48 ± 0.0 0.994 ± 0.0 0.746 ± 0.0 7.40 ± 0.1 1.96 ± 0.0 1.44 ± 0.0 0.587 ± 0.0 1.68 ± 0.0 2.00 ± 0.0 1.47 ± 0.0 0.810 ± 0.0 1.95 ± 0.0 2.42 ± 0.0 0.824 ± 0.0 1.47 ± 0.0 1.18 ± 0.0 1.16 ± 0.0

0.684 ± 0.0 1.08 ± 0.0 0.634 ± 0.0 0.497 ± 0.0 0.208 ± 0.0 0.501 ± 0.0 1.19 ± 0.0 0.895 ± 0.0 0.341 ± 0.0 2.27 ± 0.0 0.474 ± 0.0 1.52 ± 0.0 0.160 ± 0.0 0.664 ± 0.0 0.311 ± 0.0 0.479 ± 0.0 0.488 ± 0.1 1.06 ± 0.1 0.656 ± 0.0

24.5 ± 1.6 19.2 ± 1.2 5.48 ± 2.3 45.3 ± 2.8 3.50 ± 1.0 7.49 ± 0.3 6.21 ± 0.3 3.80 ± 1.3 6.27 ± 2.2 116 ± 4.1 7.82 ± 1.9 60.2 ± 1.9 10.8 ± 0.3 96.8 ± 2.8 8.53 ± 0.7 5.81 ± 1.9 5.04 ± 1.2 39.2 ± 1.5 32.1 ± 2.4

59.7 ± 1.6 40.5 ± 0.9 58.7 ± 2.9 28.7 ± 0.4 12.2 ± 0.5 5.40 ± 0.2 75.6 ± 0.7 18.6 ± 0.1 7.28 ± 0.2 70.9 ± 4.5 18.0 ± 1.6 128 ± 19.9 5.71 ± 0.4 49.6 ± 5.0 12.5 ± 0.4 50.1 ± 0.6 38.0 ± 1.6 61.8 ± 1.3 38.4 ± 0.2

375 ± 4.2 201 ± 4.8 127 ± 0.4 273 ± 2.0 188 ± 2.1 50.7 ± 1.3 297 ± 1.8 97.9 ± 1.4 110 ± 0.6 41.5 ± 1.0 95.4 ± 2.4 105 ± 0.5 37.2 ± 0.5 132 ± 4.0 20.1 ± 0.1 81.2 ± 1.8 110 ± 2.5 83.9 ± 0.9 97.3 ± 1.7

5.49 ± 2.5 35.3 ± 6.8 19.4 ± 0.3 46.9 ± 2.5 14.8 ± 1.7 30.5 ± 2.1 39.5 ± 1.8 202 ± 0.3 12.0 ± 0.1 1537 ± 4.2 15.8 ± 1.2 971 ± 4.9 66.9 ± 2.1 1186 ± 6.1 9.81 ± 0.3 23.2 ± 2.7 11.8 ± 0.2 551 ± 3.3 395 ± 2.5

28.5 ± 2.6 808 ± 0.8 41.6 ± 0.3 997 ± 7.3 1065 ± 3.8 1525 ± 6.4 844 ± 0.2 655 ± 4.7 304 ± 7.0 162 ± 0.7 339 ± 2.4 126 ± 0.8 94.6 ± 0.8 155 ± 10.5 567 ± 2.1 90.1 ± 0.3 241 ± 1.3 234 ± 1.8 66.6 ± 0.4

115 ± 14.1

1.82

101 ± 6.5

1.96

165 ± 4.9

2.59

160 ± 6.8

2.47

134 ± 10.6

2.73

62.5 ± 27.4

1.76

133 ± 9.9

2.81

125 ± 2.6

2.22

187 ± 5.3

2.95

156 ± 6.7

2.85

170 ± 8.3

2.58

141 ± 3.5

2.35

146 ± 6.6

2.31

149 ± 1.5

2.88

16.6 ± 0.8

2.83

57.3 ± 4.4

2.57

117 ± 8.7

3.00

238 ± 5.7

2.58

59.4 ± 4.7

2.19

MAT = mean annual precipitation; TOC = total organic carbon; TN = total nitrogen; TP = total phosphorus; AN = available nitrogen; AP = available phosphorous; AK = available potassium.

were taken from the headspace using gas-tight pressure-lock syringes and determined immediately. Liquid samples were taken from the sacrificed bottles. Before sampling, the bottles were shaken thoroughly for 30s with hand. Soil slurries were then divided into three subsamples for the analyses of: i) intermediate volatile fatty acids (VFAs), ii) anionic oxidants sulfate and nitrate, and iii) Fe(II) and Fe(III). The subsamples for the analysis of Fe(II) and Fe(III) were taken and immediately processed using the ferrozine reagent following the procedure described previously (Lovley and Phillips, 1987) with a slight modification (Zhang et al., 2018). About 0.5 g of soil slurry were added into a weighted tube with 4.5 ml of 0.5 M HCl under anoxic condition, mixed and shaken at a moderate speed at 30 °C for 24 h. The mixture was then centrifuged and an aliquot of the supernatant was diluted with 0.5 M HCl, mixed with ferrozine reagent prior to the measurement at 562 nm with a spectrophotometer. Total extractable Fe was measured after reduction to Fe (II) with hydroxylamine (Krumbock and Conrad, 1991). The other two subsamples were centrifuged for 10 min at 8000 rpm at 4 °C and the supernatants were transferred into Eppendorf tubes and stored frozen at −20 °C. Prior to fatty acids measurement, the frozen samples (1.5 ml) were thawed and acidified by 18.4 mmol/l H2SO4, filtered through 0.22 μm cellulose membrane and then determined by HPLC (Agilent 1260, USA) with an UV detector at 210 nm (Krumbock and Conrad, 1991).

For the analysis of sulfate and nitrate, the liquid samples (5 ml) were centrifuged second time (12,000 rpm, 10 min) and filtered through 0.45 μm membrane filters to remove residual soil particles. The concentration of anions was analyzed by ion chromatography (Metrohm 792) (Bak et al., 1991). The rates of methane production were calculated during the linear increase phase of methane production using the slope and expressed in mM CH4 d−1 (Yuan et al., 2014). The lag phase of methanogenesis was referred to the time from the beginning to the linear increasing of CH4 concentration in incubators. The depletion times of sulfate and Fe (III) were defined as the times when the concentration of sulfate was decreased to the undetectable level and when the concentration of Fe (II) was close to be constant, respectively. 3. Results 3.1. Chemical properties of soil samples The chemical properties of 19 soils were summarized in Table 1. Soil pH value ranges from 4.4 to 8.0. The soils from south China are relatively acidic, while the soils from north and middle China were neutral or slightly alkalic. Soil TOC ranges from 18 to 65 g/kg soil in paddy soils. TOC in wetland soils showed a larger variation, with the high values amounting to N200 g/kg soil in two wetlands from north China (XKW

284

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

and HBW) and the low values to 11–22 g/kg soil in three wetlands from south China. The variation of total N and P contents was in accordance with the pattern of TOC. Soils contain sufficient available P and K for supporting plant growth except three wetland soils where AP was relatively low (b10 mg/kg).

3.2. Methane production The production of CH4 displayed an initial lag followed by a rapid increase to a level without further increase due to the depletion of substrates (Fig. 1). Most soils showed a single linear increase following

Fig. 1. Methane production in 19 soils during anaerobic incubation after addition of 10 mM glucose. Shown are the total concentrations of CH4 in headspace expressed as millimoles per liter (mM) of soil slurry in incubators. The error bars indicate the standard deviations of four replicates. Soil samples were collected across different regions of China. The left panels from the dashed line represent 14 paddy field soils and the right from dashed line are 5 wetland soils. Graphs are arranged with up ones for the soils from north China and down ones for the soils from south China. The abbreviation of each sampling site is given at upper left corner in each graph: HH (Heihe), WC (Wuchang), SY (Shenyang), TJ (Tianjin), JN (Jining), QD (Qidong), CD (Chengdu), HZ (Hangzhou), YY (Yiyang), AS (Anshun), GZ (Ganzhou), DL (Dali), ZS (Zhongshan), LS (Lingshui); XKW (XingKai Lake Wetland), HBW (Haibei Alpine Wetland), THW (Tai Lake Wetland), DTW (Dongting Lake Wetland), PYW (Poyang Lake Wetland).

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

285

the initial lag phase. But in AS soil and slightly in DTW soil (Fig. 1, AS and DTW), there was a second linear increase in the later stage. Otherwise, the patterns of CH4 production were similar among the paddy (Fig. 1, left from the dashed line) and wetland soils (Fig. 1, right from the dashed line). Application of 10 mM glucose yielded approximately 30 mM CH4 (normalized to the liquid volume) in nine paddy soils and all five wetland soils (Fig. 1). The production of CH4 in other five paddy soils (HH, JN, QD, HZ, GZ) was slightly lower (25 mM CH4). These results indicate that the decomposition of glucose was close to the stoichiometric conversion of glucose to CH4 and CO2 (C6H12O6 = 3CH4 + 3CO2). Only a marginal fraction of substrate might be assimilated into microbial biomass or shunted to other metabolisms. The maximal rate (Vmax) of CH4 production in each soil was calculated during the linear increase of CH4 production. The Vmax values range from 1.8 to 3.0 mM CH4 per day (Table 1). Eleven soils showed Vmax between 2.5 and 3.0 mM CH4 per day, while three soils, all of them from north China, had the relative lower rates (1.8–2.0 mM CH4 per day). Although the pattern and the maximal rate of CH4 production did not vary substantially, the significant difference was revealed in the lag period of CH4 production among soils. Four soils showed a lag period of less than one week, eleven soils had the lag phase between one and two weeks, and the remaining four soils showed the lag phase from two to over four weeks. Most of soils with the shorter lag were from north China, while all of soils with the longer lag were from south China. When the lag periods of nineteen soils were put into a geographic map (Fig. 2A), it appears that the lag period decreased generally from south to north China.

highest content of reducible Fe(III) was detected in the ZS paddy soil and the lowest in the AS paddy soil. Both soils are from south China. It appears that the soils from south China showed a larger variation in the reducible Fe(III) content than those from north China. But unlike nitrate and sulfate, there was no obvious geographic tendency in the content of reducible Fe(III). The reduction of nitrate occurred rapidly regardless of the initial concentration. In most soils, the concentration of nitrate decreased to the undetectable level (b5 μM) one day after incubation (Fig. 4). The reduction of ferric iron also started immediately upon the addition of glucose. But unlike the total content of reducible Fe(III), the time needed for the depletion of reducible Fe(III) [estimated from the time when the concentration of Fe(II) reached its maximum] showed an obvious geographical difference among soils (Fig. 2B). In a few soils, like the HH, JN, and GZ paddy soils, it took b5 days for the depletion of reducible Fe(III). In other soils, over 10 to 15 days were needed (i.e. CD, ZS, LS and DTW soils). A general trend was that the soils from south China took a longer time than those from north China, in coincidence with the lag phase of CH4 production. Similar to ferric iron, the reduction of sulfate started immediately after anaerobic incubation in most soils. In the HH, WC, HZ, GZ and HBW soils, it took b5 days for the complete reduction of sulfate, while in the AS paddy soil and two wetland soils from south China (PYW and DTW) about 20–30 days were required (Fig. 2C). This geographical trend also coincided with the reduction of Fe(III) and the lag phase of CH4 production.

3.3. Production of volatile fatty acids

4.1. Geographic pattern of methanogenesis

Acetate, propionate and butyrate were the major intermediates during the glucose fermentation and CH4 production (Fig. 3). Acetate showed the highest transient accumulation followed by butyrate while propionate was low (b0.5 mM) in all samples. Acetate production initiated immediately after anaerobic incubation and within b10 days the concentration reached to the maxima in most soils. In four paddy soils (SY, TJ, JN, CD) and two wetland soils (HBW, THW), the maximal concentration of acetate ranged between 15 and 20 mM, while about 10–15 mM acetate was detected in other soils (Fig. 3). Butyrate accumulated transiently to a concentration of 5–10 mM in soils where the accumulation of acetate was b15 mM. In most soils, acetate and butyrate concentrations decreased to the undetectable levels three weeks after the incubation. But in two paddy soils (Fig. 3, AS and LS) and two wetland soils (Fig. 3, DTW and PYW), it took more than a month for the decomposition of these VFAs. The delayed degradation of butyrate in AS and DTW soils coincided with the second phase of linear increase of CH4 production (Fig. 1, AS and DTW). The soil samples showing the longer lag of butyrate degradation are from south China.

The present experiment shows that soils collected from different locations and vegetations shared a basic pattern of CH4 production with the maximal rate of CH4 production varying between 2 and 3 mM per day (Table 1). The production of CH4 is the final step of anaerobic decomposition of complex organic substances. Glucose was used as a substrate in the present experiment to exclude the complicated effects of upstream processes in the organic polymer degradation. Production of CH4 from glucose can be characterized by the following parameters: i) the total amount, ii) the maximal rate, and iii) the lag period. Our results revealed that the total amount and the maximal rate of CH4 production varied only moderately among different soils, roughly being 3 CH4 per glucose and 2–3 mM CH4 per day, respectively. However, the lag period of methanogenesis exhibited a explicit geographical pattern, lengthening from less than one week in north China to over four weeks in south China (Fig. 2A). Composition of intermediate products reflects the pattern of glucose fermentation. During the anaerobic degradation of glucose, the transient accumulation of acetate was the highest, amounting to 15–20 mM. This result was consistent with a previous study showing that acetate were the most significant intermediate, while propionate, butyrate/ethanol, succinate and lactate together made up b1–8% during anaerobic degradation of glucose in the European paddy soil and lake sediment (Krumbock and Conrad, 1991). The glucose fermentation via Embden-Meyerhof-Parnas (EMP) pathway typically generates two acetic acids plus two NADH and two reduced ferredoxin (Fdred) per glucose molecule. The reoxidation of NADH and Fdred generates H2 or formate which is used for anaerobic respirations or methanogenesis. A few fermenters perform homoacetogenesis, producing three acetic acids per glucose (Krumbock and Conrad, 1991). Only in two soils (TJ and CD) the accumulation of acetate was slightly higher than 20 mM, indicating that homoacetogenesis was not significant in our soils. Some clostridia perform butyrate and propionate fermentations from glucose, with the stoichiometry of one butyric or propionic acid per glucose. Butyrate and propionate produced is then degraded by secondary fermenters in syntrophic cooperation with H2/formate-scavenging methanogens (Stams and Plugge, 2009; Müller et al., 2010; Sieber et al., 2012;

3.4. Dynamics of nitrate, sulfate and ferric iron The initial concentration of nitrate was lower than 100 μM in 13 soils (Table 1). But three soils (QD, ZS and LS) showed a concentration between 200 and 550 μM and another three soils (CD, YY and DTW) showed a concentration up to N1000 μM. The soil samples with the high content of nitrate were located in south China. The initial concentration of sulfate ranged from 28 to 1525 μM in soils (Table 1). Eight soils showed a concentration of lower than 200 μM, four soils had a concentration between 200 and 340 μM and the remaining seven soils had the concentration between 600 and 1525 μM. In most soils, the concentration of sulfate were higher than that of nitrate; but in contrast to the tendency of nitrate, the initial concentration of sulfate was on the average higher in north China soils than in south soils. Six out of seven soils with high sulfate content (N600 μM) was from north China. The reducible ferric iron, which was estimated from the concentrations of ferrous iron in incubation, ranged from 17 to 240 μmol g−1 soil (Table 1). The

4. Discussion

286

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

Nazaries et al., 2013). Propionate fermentation accounted for only a marginal fraction in our soils, whereas butyrate fermentation was significant (Fig. 3). Notably, most of the south China soils showed an accumulation of butyrate up to 5–10 mM, whereas the production was lesser in the north China soils. These results suggest that the pathway of glucose fermentation also exhibits a geographical pattern, in line with the lag phase of methanogenesis. 4.2. Reduction of major oxidant ions The reduction of nitrate, sulfate and ferric iron has been considered to outcompete methanogenesis for common substrates H2 and acetate (Liesack et al., 2000; Jelen et al., 2016). Although some studies found that nitrate addition caused complete inhibition of methanogenesis (Achtnich et al., 1995a; Klüber and Conrad, 1998; Chidthaisong and Conrad, 2000; Yuan and Lu, 2009), the variation of initial nitrate concentration has little effects on CH4 production in our experiment. It is possible that the reduction of nitrate regardless of its initial concentration in soil was too rapid to exert an obvious influence. The reduction of sulfate and Fe(III), though occurring also rapidly, took a much longer time for the complete depletion of sulfate and Fe(III). Notably, the depletion time of sulfate and Fe(III) was in good coincidence with the lag phase of CH4 production (Fig. 2). The effect of sulfate and Fe(III) reduction on methanogenesis has been well documented previously (Achtnich et al., 1995a; Krylova et al., 1997). The addition of exogenous sulfate and ferrihydrite (low crystallinity and metastability) or amorphous Fe (OH)3 strongly suppressed methanogenesis (Achtnich et al., 1995a; Chidthaisong and Conrad, 2000; van Bodegom et al., 2004; Reiche et al., 2008; Hori et al., 2009). It was therefore explained that Fe(III) and sulfate reducers outcompeted methanogens by decreasing the concentration of common electron donors such as H2 below the thresholds for methanogens (Lovley et al., 1982; Achtnich et al., 1995a; Achtnich et al., 1995b). It has to be noted, however, that a low concentration of CH4 (data not shown) was detected immediately after the anaerobic incubation in most soils except PYW and DTW where CH4 was detected the fourth day after incubation. The lag phase in the present experiment referred to the time from beginning to the linear increase of CH4 concentration. However, a slow production of CH4 actually occurred in this phase. Since acetate accumulated and did not decrease during this period, it was due to hydrogenotrophic methanogenesis that contributed to the CH4 production during the early period. The concomitant occurrence of hydrogenotrophic methanogenesis (albeit slow) and the reduction of Fe(III) and sulfate indicates that competition hypothesis did not stand true during the lag phase of anaerobic incubation, in line with a previous report (Roy et al., 1997). On the other hand, the coincidence between acetate decline and the rapid increase of CH4 after the lag phase indicates that aceticlastic methanogenesis contributed to the CH4 production during the linear increase phase. Since the reduction of Fe(III) and sulfate was already complete at this stage, there should be also no competition for acetate between aceticlastic methanogens and iron and sulfate reducers. Therefore, it is unlikely that competition was the cause for the coincidence between the lag phase of methanogenesis and the depletion time of Fe(III) and sulfate observed in our soils. Glucose applied at the start presumably provided sufficient electron donors to prevent this kind of competitions. Fig. 2. Geographic patterns of the lag phase of methanogenesis (A), the depletion time of reducible ferric iron or Fe(III) (B), and the depletion time of sulfate (C). The lag phase of methanogenesis was defined as the time from the beginning to the linear increase of CH4 concentration in the headspace of incubators. The depletion time of Fe(III) and sulfate was referred to the time when the concentration of ferrous iron Fe(II) did not further increase or close to be constant and when the concentration of sulfate was decreased to the undetectable level, respectively. Abbreviations of sample sites (referred to the Legend of Fig. 1) are given in each map. Bars indicate the length of time (in days) and their colors are distinct with paddy field soils in blue and natural wetlands in red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

287

Fig. 3. The concentration (mM) of three major volatile fatty acids (VFA), acetate, propionate and butyrate, during anaerobic incubation of 19 soil samples after addition of 10 mM glucose. The error bars indicate the standard deviations of four replicates. The graph arrangement and the abbreviations of sampling sites are referred to the Legend of Fig. 1.

The finding that the lag phase of methanogenesis and the depletion time of sulfate and ferric iron were longer in south China than in north China was striking. Temperature has been considered as the most important factor shaping the geographical distribution of terrestrial biota. A survey across North American forest soils indicated that temperature also shapes the microbial diversity (Zhou et al., 2016). Even more obvious is the microbial activity in soils that in most cases is positively correlated with temperature. Previous studies on methanogenesis in paddy fields and wetland soils showed that the rate of CH4 production increased with the temperature within the range of local environments (Peng et al., 2008; Fu et al., 2015). The mean annual temperature in our sampling locations ranges from −1.2

°C in north China to 25.0 °C in south China (Zhang et al., 2018). It was expected that the south China soils might harbor a more readily activated methanogenic populations and hence a shorter lag phase of methanogenesis than the north soils. The results obtained, however, are in contrast to this prediction. Therefore, though the structure and activity of soil microbial community are believed to be positively influenced by temperature, it is unlikely that temperature at the original soils exerts a strong influence on the geographical difference in methanogenesis observed in this study. The pattern of methanogenesis is probably related to other factors like the availability of organic substrates and the nutrient conditions. Though a large difference in TOC among soils, there is no correlation

288

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

Fig. 4. The concentration of ferrous iron [Fe(II) in micromoles per gram of dry soil] and two anionic oxidants nitrate and sulfate (in micromoles per liter of soil slurry, mM) during anaerobic incubation of 19 soil samples after addition of 10 mM glucose. The error bars indicate the standard deviations of four replicates. The graph arrangement and the abbreviations of sampling sites are referred to the Legend of Fig. 1.

between CH4 production and TOC. This is probably due to the provision of sufficient substrate (glucose) in the incubations. Other nutrients, N, P, and K, appear also to be sufficient and no correlation can be established with their contents to methanogenesis (Table 1). By comparison, soil pH appears to serve as the key factor influencing the geographical pattern in methanogenesis and other reduction processes. When we plotted the lag phase of methanogenesis and the depletion time of Fe(III) and sulfate against soil pH, a clear tendency appears showing the increasing variation with a high frequency of longer lag of methanogenesis or depletion time for Fe(III) and sulfate at lower pH values (b5.5) (Fig. 5A). The statistical analysis showed a significant linear correlation between the methanogenesis lag and the depletion time of ferric iron and sulfate (Fig. 5B). This coincidence indicates that the geographical pattern of anaerobic activities is robust and related to soil pH. Acidic soils are developed vastly in south China due to the strong weathering of minerals and leaching of bases. By contrast, most of soils from north China have a neutral to slight alkalic pH. The soil pH has been identified as the most

important factor influencing the microbial community assembly and biodiversity across a large geographical distances (Fierer and Jackson, 2006; Thompson et al., 2017; Delgado-Baquerizo et al., 2018). The soil pH is also considered to regulate the balance between stochastic and deterministic processes in microbial community structuring (Tripathi et al., 2018). It is well recognized that the soil pH can not only influence soil nutrient availability but also regulate microbial metabolisms (Delgado-Baquerizo et al., 2018). The activity of most organisms is limited at pH values lower than their physiological ranges. Therefore, the geographical patterns of methanogenesis and the reduction of ferric iron and sulfate are very likely related to soil pH variation across a large distance of China paddy and wetland soils. 5. Summary In summary, our results reveal that the south China soils have a relatively longer lag phase of methanogenesis as well as the time for

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

289

References

Fig. 5. The plot showing the lag phase of methanogenesis and the depletion time of Fe(III) and sulfate against soil pH (A) and the linear correlation between the depletion time of Fe (III) and sulfate and the lag phase of methanogenesis (B).

complete reduction of ferric iron and sulfate than the north China soils. This geographical difference is likely related to the soil pH variation. This pattern suggests that the production and emission of CH4 may boost faster in north China than south China under similar temperature scenarios. Given that the global warming effect is predicted to be more severe at high than low latitudes, our results suggest that the north China soils may yield a stronger feedback response of CH4 emissions than the south China soils. Conflict of interest The authors declare no conflict of interest associated with this work. Author contributions XH and YL devised the study, XH performed the experiments. XH and SJ analyzed the data. All authors contributed to data interpretation and writing of the manuscript. Acknowledgment This research was supported by the National Key Research and Development Program of China (2016YFD0200306) and the National Natural Science Foundation of China (41630857).

Achtnich, C., Bak, F., Conrad, R., 1995a. Competition for electron-donors among nitrate reducers, ferric iron reducers, sulfate reducers, and methanogens in anoxic paddy soil. Biol. Fertil. Soils 19 (1), 65–72. Achtnich, C., Schuhmann, A., Wind, T., Conrad, R., 1995b. Role of interspecies H2 transfer to sulfate and ferric iron-reducing bacteria in acetate consumption in anoxic paddy soil. FEMS Microbiol. Ecol. 16 (1), 61–69. https://doi.org/10.1111/j.1574-6941.1995. tb00269.x. Bak, F., Scheff, G., Jansen, K.-H., 1991. A Rapid and Sensitive Ion Chromatographic Technique for the Determination of Sulfate and Sulfate Reduction Rates in Freshwater Lake Sediments. Bao, S.D., 2000. Soil and Agricultural Chemistry Analysis. Agriculture Publication, Beijing, pp. 355–356. Bhatia, A., Jain, N., Pathak, H., 2013. Methane and nitrous oxide emissions from Indian rice paddies, agricultural soils and crop residue burning. Greenh. Gases Sci. Technol. 3 (3), 196–211. van Bodegom, P.M., Scholten, J.C.M., Stams, A.J.M., 2004. Direct inhibition of methanogenesis by ferric iron. FEMS Microbiol. Ecol. 49 (2), 261–268. Bridgham, S.D., Cadillo-Quiroz, H., Keller, J.K., Zhuang, Q.L., 2013. Methane emissions from wetlands: biogeochemical, microbial, and modeling perspectives from local to global scales. Glob. Chang. Biol. 19 (5), 1325–1346. Chen, H., Zhu, Q.A., Peng, C.H., Wu, N., Wang, Y.F., Fang, X.Q., et al., 2013. Methane emissions from rice paddies natural wetlands, lakes in China: synthesis new estimate. Glob. Chang. Biol. 19 (1), 19–32. Chidthaisong, A., Conrad, R., 2000. Turnover of glucose and acetate coupled to reduction of nitrate, ferric iron and sulfate and to methanogenesis in anoxic rice field soil. FEMS Microbiol. Ecol. 31 (1), 73–86. Delgado-Baquerizo, M., Oliverio, A.M., Brewer, T.E., Benavent-González, A., Eldridge, D.J., Bardgett, R.D., et al., 2018. A global atlas of the dominant bacteria found in soil. Science 359 (6373), 320–325. https://doi.org/10.1126/science.aap9516. Drake, H.L., Horn, M.A., Wust, P.K., 2009. Intermediary ecosystem metabolism as a main driver of methanogenesis in acidic wetland soil. Environ. Microbiol. Rep. 1 (5), 307–318. Fierer, N., Jackson, R.B., 2006. The diversity and biogeography of soil bacterial communities. Proc. Natl. Acad. Sci. U. S. A. 103 (3), 626–631. https://doi.org/10.1073/ pnas.0507535103. Fu, L., Song, T., Lu, Y., 2015. Snapshot of methanogen sensitivity to temperature in Zoige wetland from Tibetan plateau. Front. Microbiol. 6. https://doi.org/10.3389/ fmicb.2015.00131. Hori, T., Müller, A., Igarashi, Y., Conrad, R., Friedrich, M.W., 2009. Identification of ironreducing microorganisms in anoxic rice paddy soil by 13C-acetate probing. ISME J. 4, 267. https://doi.org/10.1038/ismej.2009.100. Jelen, B.I., Giovannelli, D., Falkowski, P.G., 2016. The role of microbial electron transfer in the coevolution of the biosphere and geosphere. Annu. Rev. Microbiol. 70 (70), 45-+. King, G.M., 1984. Utilization of hydrogen, acetate, and noncompetitive substrates by methanogenic bacteria in marine-sediments. Geomicrobiol J. 3 (4), 275–306. Klüber, H.D., Conrad, R., 1998. Effects of nitrate, nitrite, NO and N2O on methanogenesis and other redox processes in anoxic rice field soil. FEMS Microbiol. Ecol. 25 (3), 301–318. https://doi.org/10.1111/j.1574-6941.1998.tb00482.x. Krumbock, M., Conrad, R., 1991. Metabolism of position-labeled glucose in anoxic methanogenic paddy soil and lake sediment. FEMS Microbiol. Ecol. 85 (3), 247–256. Krylova, N.I., Janssen, P.H., Conrad, R., 1997. Turnover of propionate in methanogenic paddy soil. FEMS Microbiol. Ecol. 23 (2), 107–117. Liesack, W., Schnell, S., Revsbech, N.P., 2000. Microbiology of flooded rice paddies. FEMS Microbiol. Rev. 24 (5), 625–645. Lovley, D.R., Goodwin, S., 1988. Hydrogen concentrations as an indicator of the predominant terminal electron-accepting reactions in aquatic sediments. Geochim. Cosmochim. Acta 52 (12), 2993–3003. Lovley, D.R., Phillips, E.J.P., 1987. Competitive mechanisms for inhibition of sulfate reduction and methane production in the zone of ferric iron reduction in sediments. Appl. Environ. Microbiol. 53 (11), 2636–2641. Lovley, D.R., Dwyer, D.F., Klug, M.J., 1982. Kinetic-analysis of competition between sulfate reducers and methanogens for hydrogen in sediments. Appl. Environ. Microbiol. 43 (6), 1373–1379. Megonigal, P., Hines, M., Visscher, P., 2003. Anaerobic Metabolism: Linkages to Trace Gases and Aerobic Processes. Menon, S., Denman, K.L., Brasseur, G., Chidthaisong, A., Ciais, P., Cox, P.M., et al., 2007. Couplings Between Changes in the Climate System and Biogeochemistry. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). Müller, N., Worm, P., Schink, B., Stams Alfons, J.M., Plugge Caroline, M., 2010. Syntrophic butyrate and propionate oxidation processes: from genomes to reaction mechanisms. Environ. Microbiol. Rep. 2 (4), 489–499. https://doi.org/10.1111/j.17582229.2010.00147.x. Nazaries, L., Murrell, J.C., Millard, P., Baggs, L., Singh Brajesh, K., 2013. Methane, microbes and models: fundamental understanding of the soil methane cycle for future predictions. Environ. Microbiol. 15 (9), 2395–2417. https://doi.org/10.1111/14622920.12149. Nelson, M.B., Martiny, A.C., Martiny, J.B.H., 2016. Global biogeography of microbial nitrogen-cycling traits in soil. Proc. Natl. Acad. Sci. 113 (29), 8033–8040. https:// doi.org/10.1073/pnas.1601070113. Peng, J., Lu, Z., Rui, J., Lu, Y., 2008. Dynamics of the methanogenic archaeal community during plant residue decomposition in an anoxic rice field soil. Appl. Environ. Microbiol. 74 (9), 2894–2901. https://doi.org/10.1128/AEM.00070-08. Peters, V., Conrad, R., 1996. Sequential reduction processes and initiation of CH4 production upon flooding of oxic upland soils. Soil Biol. Biochem. 28 (3), 371–382.

290

X. Hao et al. / Science of the Total Environment 651 (2019) 281–290

Qu, D., Ratering, S., Schnell, S., 2004. Microbial reduction of weakly crystalline iron (III) oxides and suppression of methanogenesis in paddy soil. Bull. Environ. Contam. Toxicol. 72 (6), 1172–1181. Reiche, M., Torburg, G., Kusel, K., 2008. Competition of Fe(III) reduction and methanogenesis in an acidic fen. FEMS Microbiol. Ecol. 65 (1), 88–101. Roy, R., Kluber, H.D., Conrad, R., 1997. Early initiation of methane production in anoxic rice soil despite the presence of oxidants. FEMS Microbiol. Ecol. 24 (4), 311–320. Saunois, M., Bousquet, P., Poulter, B., Peregon, A., Ciais, P., Canadell, J.G., et al., 2016. The global methane budget 2000–2012. Earth Syst. Sci. Data 8 (2), 697–751. https://doi. org/10.5194/essd-8-697-2016. Schaefer, H., Fletcher, S.E.M., Veidt, C., Lassey, K.R., Brailsford, G.W., Bromley, T.M., et al., 2016. A 21st-century shift from fossil-fuel to biogenic methane emissions indicated by (CH4)-C-13. Science 352 (6281), 80–84. https://doi.org/10.1126/science.aad2705. Sieber, J.R., McInerney, M.J., Gunsalus, R.P., 2012. Genomic insights into syntrophy: the paradigm for anaerobic metabolic cooperation. Annu. Rev. Microbiol. 66 (1), 429–452. https://doi.org/10.1146/annurev-micro-090110-102844. Stams, A.J.M., Plugge, C.M., 2009. Electron transfer in syntrophic communities of anaerobic bacteria and archaea. Nat. Rev. Microbiol. 7, 568. https://doi.org/10.1038/ nrmicro2166. Thauer, R.K., Kaster, A.K., Seedorf, H., Buckel, W., Hedderich, R., 2008. Methanogenic archaea: ecologically relevant differences in energy conservation. Nat. Rev. Microbiol. 6 (8), 579–591. Thompson, L.R., Sanders, J.G., McDonald, D., Amir, A., Ladau, J., Locey, K.J., et al., 2017. A communal catalogue reveals Earth's multiscale microbial diversity. Nature https:// doi.org/10.1038/nature24621. Tripathi, B.M., Stegen, J.C., Kim, M., Dong, K., Adams, J.M., Lee, Y.K., 2018. Soil pH mediates the balance between stochastic and deterministic assembly of bacteria. ISME J. 12 (4), 1072–1083. https://doi.org/10.1038/s41396-018-0082-4. Wen, X., Yang, S., Horn, F., Winkel, M., Wagner, D., Liebner, S., 2017. Global biogeographic analysis of methanogenic archaea identifies community-shaping environmental factors of natural environments. Front. Microbiol. 8 (1339). https://doi.org/10.3389/ fmicb.2017.01339.

Winfrey, M.R., Zeikus, J.G., 1977. Effect of sulfate on carbon and electron flow during microbial methanogenesis in freshwater sediments. Appl. Environ. Microbiol. 33 (2), 275–281. Yan, X., Ohara, T., Akimoto, H., 2003. Development of region-specific emission factors and estimation of methane emission from rice fields in the East, Southeast and South Asian countries. Glob. Chang. Biol. 9 (2), 237–254. https://doi.org/10.1046/j.13652486.2003.00564.x. Yuan, Q., Lu, Y., 2009. Response of methanogenic archaeal community to nitrate addition in rice field soil. Environ. Microbiol. Rep. 1 (5), 362–369. https://doi.org/10.1111/ j.1758-2229.2009.00065.x. Yuan, Q., Pump, J., Conrad, R., 2014. Straw application in paddy soil enhances methane production also from other carbon sources. Biogeosciences 11 (2), 237–246. Zehnder, A.J.B., 1988. Geochemistry and biogeo-chemistry of anaerobic habitats. Biol. Anaerobic Microorg. 1–38. Zhang, J., Lu, Y., 2016. Conductive Fe3O4 nanoparticles accelerate syntrophic methane production from butyrate oxidation in two different lake sediments. Front. Microbiol. 7 (1316). https://doi.org/10.3389/fmicb.2016.01316. Zhang, B.W., Tian, H.Q., Ren, W., Tao, B., Lu, C.Q., Yang, J., et al., 2016. Methane emissions from global rice fields: magnitude, spatiotemporal patterns, and environmental controls. Glob. Biogeochem. Cycles 30 (9), 1246–1263. Zhang, J., Jiao, S., Lu, Y.H., 2018. Biogeographic distribution of bacterial, archaeal and methanogenic communities and their associations with methanogenic capacity in Chinese wetlands. Sci. Total Environ. 622, 664–675. Zhou, J., Deng, Y., Shen, L., Wen, C., Yan, Q., Ning, D., et al., 2016. Temperature mediates continental-scale diversity of microbes in forest soils. Nat. Commun. 7. https://doi. org/10.1038/ncomms12083. Zu, Q., Zhong, L., Deng, Y., Shi, Y., Wang, B., Jia, Z., et al., 2016. Geographical distribution of methanogenic archaea in nine representative paddy soils in China. Front. Microbiol. 7 (1447). https://doi.org/10.3389/fmicb.2016.01447.