Soil carbon changes in paddy fields amended with fly ash

Agriculture, Ecosystems and Environment 245 (2017) 11–21

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee

Soil carbon changes in paddy fields amended with fly ash a,b

a,⁎

b

b

a

Sang-Sun Lim , Woo-Jung Choi , Scott X. Chang , Muhammad A. Arshad , Kwang-Sik Yoon , ⁎ Han-Yong Kimc, a b c

MARK

Department of Rural and Biosystems Engineering, Chonnam National University, Gwangju, 61186, Republic of Korea Department of Renewable Resources, University of Alberta, Edmonton, Alberta, T6G 2E3, Canada Department of Applied Plant Science, Chonnam National University, Gwangju, 61186, Republic of Korea

A R T I C L E I N F O

A B S T R A C T

Keywords: Compost Green manure Mineral amendment Rice grain yield Soil carbon sequestration Synthetic fertilizer

Increasing soil carbon (C) sequestration in the agricultural sector is an important strategy for mitigating climate change; however, conventional best management practices such as crop residue retention and organic fertilizer application do not always increase soil C content due to C loss by cultivation. In this context, application of finetextured minerals such as coal fly ash (FA) may be effective in increasing soil C sequestration by enhancing plant biomass production and protecting soil C from being lost. We conducted a three-year field experiment in a paddy field with three levels of FA application (0, 5, and 10% by soil weight) in combination with the following four nitrogen (N) treatments: no input, and applications of urea, pig manure compost (compost) and hairy vetch (Vicia Villosa Roth.) green manure (vetch). Across the three seasons, rice grain yield was in the order of vetch = urea > compost > no input, reflecting the effect of N availability in each treatment. Application of FA (particularly at 10%) reduced the total rice plant biomass by hampering tillering. However, FA application did not reduce grain yield due to increased individual grain weight. In spite of decreased rice residue incorporation into the soil, FA application increased the soil C content at the end of the third season regardless of the N source, driven by reduced soil C loss. We conclude that the application of mineral soil amendments such as FA is effective in enhancing soil C sequestration without decreasing rice yield in paddy fields.

1. Introduction Agricultural activities including land-use change and intensive cultivation have substantially contributed to the increase of greenhouse gas concentrations in the atmosphere (Lal, 2004). However, if best management practices (BMPs) are adopted, agricultural soils may become an effective carbon (C) sink as the C content of the soils is below the C saturation point, an equilibrium C content at which no further C can be sequestered with time under a steady C input (Lal, 2004; West and Six, 2007). Particularly, fine-textured paddy soils are considered to be more effective for C sequestration, as compared with upland soils, due to the slow decomposition of organic C under submerged or anaerobic conditions during the growing season (Sahrawat, 2004; Rui and Zhang, 2010). Commonly used BMPs to increase C sequestration in soils include minimum or no tillage, cover cropping, addition of organic amendments, balanced fertilization, and rotational cropping (Lal, 2004; Yan et al., 2007; Tian et al., 2015). However, soil C sequestration relying on the traditional BMPs is not always efficient due to loss of applied organic C as well as native soil C loss as a result of soil disturbance by

⁎

cultivation (Yan et al., 2007; Tian et al., 2015). For example, balanced fertilization with synthetic fertilizer may produce more rice plant residues but residue incorporation may not necessarily enhance soil C sequestration as rice residues are readily decomposed during cultivation (Viswanath et al., 2010). Among organic nutrient sources such as green manure and livestock manure compost, green manure application may not directly contribute to soil C sequestration due to its faster rate of decomposition (Lim and Choi, 2014; Park et al., 2015). Application of livestock manure compost which has recalcitrant C can enhance soil C sequestration; however, compost application may reduce rice yield due to its low nutrient availability (Yun et al., 2011). In this context, the application of mineral amendments such as coal fly ash (FA) that have a high specific area with organic fertilizers as a source of nutrients can be effective to enhance soil C sequestration via soil C stabilization (Six et al., 2004; Jastrow et al., 2007). Fly ash is a byproduct from coal power plants, and it is estimated that 750 million tons of FA is generated per annum on a global basis (Blissett and Rowson, 2012). Due to the large specific area of the siltsized particles, FA provides sites for the formation of organo-mineral complexes (Jala and Goyal, 2006). In addition, FA can fix CO2 produced

Corresponding authors at: Chonnam National University, Gwangju, 61186, Republic of Korea. E-mail addresses: [email protected] (W.-J. Choi), [email protected] (H.-Y. Kim).

http://dx.doi.org/10.1016/j.agee.2017.03.027 Received 8 June 2016; Received in revised form 28 March 2017; Accepted 30 March 2017 0167-8809/ © 2017 Elsevier B.V. All rights reserved.

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

company. For green manure application, hairy vetch (Vicia Villosa Roth.) (vetch) was provided by a research station of Rural Development Administration (RDA) of Korea. The compost and vetch were freezedried, stored in a refrigerator, and used for the three-year field experiment. In each year, the compost (10 kg in dry weight) was passed through a 4-mm sieve and vetch (10 kg) was chopped to < 4 mm. A portion (100 g) of the samples was further ground to < 2 mm and analyzed for chemical properties following the same methods as soil analysis (Table 1). Additionally, chemical stability degree, the ratio of acid hydrolysable organic matter to total organic matter, was determined by a modified Klason lignin method (López et al., 2010). The chemical properties of the compost and vetch did not differ among years, and thus the average values are reported. Although the C/N ratio of the compost (10.7) and vetch (11.5) was not significantly different, the chemical stability degree was much higher for the compost (27.3%) than for the vetch (2.2%), indicating that the compost was more recalcitrant to microbial decomposition than the vetch (Table 1). Fly ash (ca. 500 kg) was obtained from a coal power plant at Hadong, Korea. A portion (20 g) of the FA sample was oven-dried at 105 °C and analyzed for pH, EC, total C and N concentration, and particle size distribution using the same methods for soil analysis. Elemental composition was determined using X-ray fluorescence spectrometry (S4 PIONEER, Bruker, Germany); water extractable metals concentration with ICP-AES (Optima-7000DV, PerkinElmer, Boston, USA) after extraction with distilled water; and NH4OAc extractable metals concentration with ICP-AES (Optima-7000DV, PerkinElmer, Boston, USA) after extraction with 1 mol L−1 ammonium acetate. The FA used in this study had typical properties of those used in other studies including high pH, EC, and CaO content (e.g., Pandey and Singh, 2010), and some of the properties of the FA were reported in our previous studies (Lim et al., 2012b; Lee et al., 2014; Lim and Choi, 2014). Briefly, the pH and EC were 11.7 and 1.55 dS m−1, respectively; CaO and MgO contents were 7.0 and 2.5%, respectively; and total C content was 24.2 g kg−1. The FA was mainly composed of silt-sized (0.002–0.05 mm) particles (75.4%) followed by sand (>0.05 mm, 22.7%), with only 1.9% clay (< 0.002 mm), and the concentrations of NH4OAc-extractable arsenic (As) (6.8 mg kg−1) and boron (B) (95.1 mg kg−1) were higher than those of other elements such as copper and zinc (< 1.2 mg kg−1) (Lim and Choi, 2014).

from soil respiration via carbonation of calcium (Ca) and magnesium (Mg), (Ca2+ or Mg2+) + CO32− → CaCO3 or MgCO3, that are present in FA, thereby decreasing CO2 emission while increasing soil C content (Lim et al., 2012a; Lim and Choi, 2014). However, it has not been verified yet if FA application can increase soil C sequestration in the field with growing plants. Lack of such information limits the practical use of FA for enhancing soil C sequestration. Originally, FA has been considered as a soil amendment for acidic infertile soils due to its high pH and the macro- and micro-elements it contains (Basu et al., 2009; Jala and Goyal, 2006). Many studies reported that FA application at a low rate (e.g., < 12% w/w) improved rice growth (Basu et al., 2009; Jala and Goyal, 2006 and references cited therein) without increasing heavy metal concentrations in the rice plant (Nayak et al., 2015). Therefore, FA-induced reduction in soil C loss together with increased rice growth may enhance soil C sequestration in paddy soils, particularly when the traditional BMPs such as plant residue retention and application of organic input (e.g., compost or green manure) are combined. In this study, we addressed the questions 1) if FA as a mineral amendment increases soil C sequestration in rice paddy?; 2) if so, what is the mechanisms (via increased rice growth vs. reduced soil C loss) of the increased soil C sequestration?; and 3) how does the effect of FA on soil C vary with nitrogen (N) sources such as synthetic fertilizer, livestock manure compost, and green manure that are co-applied with FA? To answer the questions, the effect of FA application on rice growth (total biomass production, grain yield, and parameters of growth and yield components) and soil C content was investigated over three rice growing seasons. We hypothesized that FA can enhance C sequestration in the soil via reducing gaseous C loss and/or increasing rice plant biomass and such effect of FA on soil C sequestration would be different among the typical N sources including synthetic fertilizer, green manure, and livestock manure compost due to different N availability of the N sources and contrasting decomposability of the organic matter that is added with livestock manure compost and green manure. 2. Materials and methods 2.1. Study site and soil This study was conducted in an experimental rice paddy field (126°53′E, 35°10′N, 33 m above sea level) at Chonnam National University, Gwangju, Korea from 2010 to 2012. This area belongs to a typical East Asian temperate monsoon climate system with an annual mean temperature of 13.8 °C and precipitation of 1391 mm over the past 30 years. Weather data were collected from the Gwangju meteorological station in the vicinity (200 m away) of the studied paddy field. The cumulative precipitation and solar radiation, and daily mean temperature during the rice growing seasons from June to October differed with years; e.g., higher temperature in 2010 than in 2011 and 2012 (Fig. 1). In 2012, two typhoons (Tembin and Bolaven) struck this area at the heading and grain-filling stages of rice plant growth (Fig. 1). The soil was classified as an Inceptisol (coarse loamy, mixed, mesic family of Fluvaquetic Endoaquepts) in the USDA Soil Taxonomy (RDA, 2000). Surface (0–20 cm) soil samples were collected from ten randomly located points within the field. The soils were air-dried, passed through a 2-mm sieve, and analyzed for particle-size distribution and chemical properties including pH, electrical conductivity (EC), total C, total N, total P, and available P (Table 1). The detailed analytical procedures are described in the Supplementary data.

2.3. Field experiment A total of thirty-six plots (plot size: 1 m × 1 m) were established for four N sources (no input (CK), urea, compost, and vetch) and three FA application rates (0, 5 and 10% w/w, coded as FA0, FA5, and FA10, respectively) in a split-plot design arranged in 3 blocks (Table S1). The N source was the main-plot treatment and the FA application rate was the split-plot treatment. Twelve plots (four N sources × three FA rates) were laid out in each block, and plots were spaced 1 m apart and each plot was confined by inserting flexible plastic barriers (35 cm in height) into the plow pan layer of the soil (around 20 cm deep) to minimize cross-contamination among plots. Ten days before the transplanting of rice (Oryza sativa L., cv. Ilmybyeo, subsp. japonica) in the first year (2010), FA was applied to the surface of each plot and manually mixed with the top (0–20 cm) soil. The FA application rate was determined as a percentage of FA to the dry weight of the top soil, and thus 5 and 10% are equivalent to approximately 10.7 and 21.4 kg m−2 (Table S1). The application of FA at 5 and 10% supplied 260 and 520 g C m−2, respectively, in the form of black C and it was calculated that the addition of black C (mostly associated with silt-sized particles) at those two application rates increased C concentration of silt-sized particles of the bulk soils by 1.4 and 2.8 g C kg−1, respectively. No FA was applied in 2011 and 2012. Nutrients were applied 3 days before transplanting each year. Fused phosphate (4.5 g P2O5 m−2) and KCl (4 g K2O m−2) were mixed

2.2. N sources and fly ash Chemical grade urea, fused phosphate and KCl were used as synthetic fertilizers. Pig manure compost (compost) that was produced by composting pig manure with sawdust as a bulking agent for approximately 2 months was purchased from a compost manufacturing 12

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Fig. 1. Daily mean precipitation, air temperature, and solar radiation during the three rice growing seasons: (a) 2010 (year 1), (b) 2011 (year 2), and (c) 2012 (year 3). Rice growth stage and agricultural management (fertilization and harvest) are depicted at the top of the figure. Two typhoons that occurred in 2012 are also indicated. Cumulative precipitation (Pcum), daily mean temperature (Tmean), and cumulative solar radiation (Radiationcum) are provided.

2.4. Sampling and analyses

thoroughly with the top soil (0–20 cm) in all plots. The urea, compost, and vetch were used as N sources and applied at the standard N rate (11 g N m−2) for paddy rice cultivation in South Korea (Yun et al., 2011). In the compost and vetch treatments, the entire amount of the organic fertilizer was added as a basal application only, and urea was split-applied over three times: as a basal fertilizer (5.5 g N m−2, 3 days before transplanting), the first top-dressing at tillering (3.3 g N m−2, 21 days after transplanting), and the second top-dressing at panicle initiation stage (2.2 g N m−2, 50 days after transplanting) following conventional fertilization practices of farms in the study area. Between June 2 and 6 of each year, three 30-day old seedlings (referred to as a hill) of rice were manually transplanted into each plot with a spacing of 15 cm within a row and 30 cm between rows (a total of 21 hills per plot). Rice seedlings were also transplanted in the remaining space in the field to create uniform canopy conditions. The plots were maintained under waterlogged (3–5 cm of water above the soil surface) conditions throughout the growing season by irrigation except for a 7-day summer drainage in mid-July until 120 days after transplanting.

At harvest, three hills were randomly selected in each plot and the aboveground plant part was harvested at the ground level. To collect as much belowground plant parts (root) as possible, the soil around the harvested hill was excavated intact, extending 7.5 cm long within the row, 15 cm long between rows (representing an area 15 cm long within the row and 30 cm long between rows, with the hill as the center) and 20 cm deep, with a shovel. The excavated soil was placed on a 2-mm sieve and washed with running water and all roots on the sieve were collected. The collected rice plants were brought to the laboratory, washed with distilled water, and ovendried at 60 °C to a constant weight and used for C analysis. The remaining rice plants in the plots were harvested, and after removing rice grains, the rice straw was chopped (< 5 cm) and returned to the corresponding plots. Before the commencement of this experiment, rice straw was rarely returned to the field due to the demand on rice straw for use as a bedding material and fodder in the livestock farms. 13

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

The amount of C lost during the three rice cropping seasons was calculated using the following equation:

Table 1 Properties of soil, pig manure compost (compost), and green manure hairy vetch (vetch) used in the experiment. Property

Soil

Compost

Vetch

pH

5.76 (0.07) 0.07 (0.01) 10.1 (0.1)

7.38 (0.04)

NDa

10.80 (0.10)

ND

295.0 (3.1)

Electrical conductivity (dS m−1) Total C (g C kg

−1

)

Total N (g N kg−1) C/N ratio Total P (g P kg−1) Available P (mg P2O5 kg−1) Cation exchange capacity (cmol kg−1) Chemical stability degree (%) Texture Particle size distribution (%)

Clay Silt Sand

CLOSS = SCi + CFA + CNS + CPR − SCf where SCf and SCi are final (at year 3 harvest) and initial (year 0) soil C content, respectively, CFA is C added via FA application, CNS is C added via N source, CPR is plant residue (straw and root) C incorporated into the soil at years 1 and 2, and CLOSS is the amount of C lost. It was assumed that other C inputs from root exudates, dissolved C in rainfall and irrigation water, and algal biomass in the paddy surface were not different between the treatments. As rice plant residues are readily decomposable (Viswanath et al., 2010), the residues mixed with the soils in years 1 and 2 were assumed to be decomposed sufficiently to be incorporated into the soil C pool in year 3 when the final soil sampling was conducted.

0.95 (0.03) 10.6 (0.2) 0.65 (0.02) 12.8 (0.7) 14.8 (0.5)

27.5 (0.2)

401.3 (3.8) 34.9 (1.0)

10.7 (0.3) 10.0 (0.1)

11.5 (0.2) 5.0 (0.2)

NAb NA

NA NA

NA Loam

27.3 (3.4) NA

2.2 (0.4) NA

2.6. Statistical analysis

18.5 (0.1) 48.3 (0.2) 33.2 (0.1)

NA NA NA

NA NA NA

For statistical analysis, data were tested for homogeneity of variance and normality of distribution. Logarithmic transformation was performed on grain and root dry matter and root C concentration to bring the data sets to normal distribution and homogeneous variance, and back-transformed data are reported. The following linear mixed model was used to estimate the effects of N source, FA application rate, and year, in terms of their dependence and interactions, on rice plant biomass, grain yield, and rice growth and yield parameters, and soil C content:

Values for soil are means with the standard errors of triplicate measurement in parentheses, and for compost and vetch, the values are means with the standard errors in parentheses of three experimental years. a ND: not determined. b NA: not applicable.

Zijkl = μ + Bi + Yj + (BY)ij + Nk + (BN)ik + (YN)jk + (BYN)ijk + Fl + (BF)il + (YF)jl + (NF)kl + (BYF)ijl + (BNF)ikl + (YNF)jkl + (BYNF)ijkl + εijkl

Soil samples were collected before the experiment (year 0) and after rice harvest in 2010 (year 1), 2011 (year 2), and 2012 (year 3). Around 3 kg of soil samples (0–20 cm) were collected from five points in each plot using a soil augur and composited to represent the plot. The soils from each plot were mixed thoroughly and a sub sample (∼300 g) was taken and the rest of the soil was returned to the plot. The soil samples were then brought to the laboratory, air-dried and passed through a 2mm sieve. At the same time as the above sampling, soil core (500 cm3) samples (n = 3) were also collected and measured for soil bulk density. The rice plants were separated into grain, straw (including green and dead leaves, leaf sheath and culm, and rootstock), and root. Rice growth and yield parameters including grain yield, number of tillers, number of panicles, number of grains, and individual grain weight were determined for the three hills sampled per plot. Yield and the number of tillers were scaled up to per square meter based on the planting density (21 hills per square meter), and the number of panicles per tiller and the number of grains per panicle were calculated. The plant samples were chopped with a mechanical blender and homogenized. Sub-samples of plant and soil were ground with a ball mill (MM200, Retsch Gmbh, Hann, Germany) to fine powder and analyzed for total C concentration. The changes in soil C distribution of soil particles before (year 0) and after (year 3) the experiment were determined by analyzing C concentration in organic particles, coarse sand (0.5–2 mm), medium sand (0.25–0.5 mm), fine sand (0.05–k0.25 mm), silt (0.002–0.05 mm) and clay (< 0.002 mm). The detailed procedures of particle size separation are provided in the Supplementary data. Soil samples collected at year 3 were analyzed for soil pH.

where Zijkl is the response variable (such as biomass, grain yield, and soil C content) at block i (= 1, 2, 3), year j (=1, 2, 3), N source k (=1, 2, 3, 4), and FA application rate l (=1, 2, 3); μ is the overall mean and εijkl is the error term; and B, Y, N, and F are the block, year, N source, and FA application rate effects, respectively. All experimental variables and their interactions were treated as fixed effects, while block and its interactions were treated as random effects. The model was fitted using the restricted maximum likelihood (REML) procedure in SAS® Mixed to estimate the means and standard errors for each combination of N source and FA application rate. When the treatment effects were significant, the means were separated by the Duncan’s multiple range tests. The level of significance for all statistical tests was set α = 0.05. 3. Results 3.1. Total rice plant biomass, grain yield, and residue biomass C Total rice plant biomass was affected by year (P < 0.001), N source (P < 0.001), and FA application rate (P = 0.020), and there was also interaction (P = 0.002) between N source and year (Table 2). The mean total biomass averaged across the N sources and FA application rates was greatest (P < 0.001) in year 1 than those in years 2 and 3 (Fig. S1 for data of each year). Across the three rice seasons, the total rice plant biomass was higher (P < 0.001) in urea and vetch than in compost and CK treatments (Fig. 2a). The effect of FA application on total biomass differed with application rate; FA5 did not affect total biomass, but FA10 decreased (P = 0.020) the total biomass by 7.1% on average across N sources through three years (Table 2; Fig. 2a). When the total biomass of the four N sources treatments were pooled, similar patterns were also found (Fig. 3a). Such decreases in total rice plant biomass by FA10 were largely driven by reduced straw biomass (P = 0.017) (Table 2). The responses (P < 0.001) of grain yield to N source in three rice seasons were similar to those of total rice plant biomass; across three

2.5. Calculation of soil C content and loss The C content of the bulk soil (0–20 cm) was expressed as the amount of C per square meter (g C m−2) using the mean bulk density of the soil (ca. 1.1 g cm−3). For particle-size fractions, the C concentration in each fraction was expressed as the amount of C per unit mass of the bulk soil (g C kg−1 bulk soil) as this better reflects changes in C content in each fraction than the C concentration of each fraction due to alteration of particle size fractions by silt-enriched FA application. 14

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Table 2 The P values of the analysis of variance (ANOVA) on rice plant biomass, grain yield, and rice growth and yield parameters as affected by N source (N) (no input, urea, pig manure compost, and hairy vetch green manure) and fly ash application rate (FA) (0, 5, and 10% by soil weight) across three cropping years (Y). Effect

Y N FA Y×N Y × FA N × FA Y × N × FA

Total biomass

< 0.001 < 0.001 0.020 0.002 0.091 0.821 0.733

Plant residue biomass

Grain yield

Straw

Root

Straw + Root

< 0.001 < 0.001 0.017 0.004 0.221 0.962 0.767

0.007 < 0.001 0.210 0.005 0.241 0.102 0.963

< 0.001 < 0.001 0.031 0.010 0.182 0.812 0.722

< 0.001 < 0.001 0.167 < 0.001 0.367 0.231 0.489

Growth and yield parameters Number of tillers m−2

Number of panicles tiller−1

Number of grains panicle−1

Individual grain weight

< 0.001 < 0.001 0.047 0.010 0.823 0.634 0.567

< 0.001 < 0.001 0.101 0.017 0.903 0.633 0.556

< 0.001 0.340 0.843 < 0.001 0.667 0.084 0.071

< 0.001 0.403 0.041 0.933 0.624 0.463 0.934

Fig. 2. Three-year mean data of (a) total plant biomass, (b) grain yield, and (c) plant residues (root and shoot) C of rice (oven-dry basis) grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). Vertical bars are standard errors of the mean (error bars are often too small to be visible). Right-side panels are means values across N source and FA application rate (denoted as All), for each N source across FA application rate (N source), and for each FA application rate across N sources (FA application rate). Numerical values on the panels are percentage change relative to CK or FA0. Different lowercase letters denote significant (P < 0.05) difference between either N source or FA treatment. Detailed treatments are provided in Table S1, and the results of statistical analysis are presented in Table 2. Annual data are provided in Supporting information (Figs. S1–3).

15

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

source, FA application rate, and their interactions (Fig. 5). At year 1 when plant residue (i.e., straw) was not incorporated into the soils at the previous year’s harvest, the soil C content declined except for the compost treatment as compared to the initial soil (year 0) (Fig. 4). When FA was not applied, the soil C content in the CK, urea, and vetch treatments at year 3 was lowered (by 3–14%) compared to year 0; whereas compost application increased the soil C content by 14.3% (Fig. 6a). Application of FA increased soil C content in the order of compost > vetch > urea > CK across the three rice seasons regardless of the FA application rate (Fig. 6a). When the black C contained in FA was excluded, the soil C content in the CK decreased by 3–6% regardless of the FA application rate in year 3 (Fig. 6b). In the urea treatment, the soil C content increased for FA5 (by 13.6%) but not for FA10. Co-application of FA with compost achieved the largest increment of soil C content regardless of FA application rate (by ca. 27% in year 3 as compared to year 0). When FA was co-applied with vetch, the soil C content increased by 15–18% in year 3 as compared to year 0 (Fig. 6b). Therefore, the greatest soil C increment during the rice cultivation period was achieved by co-application of compost with FA10. When the soil C content of the four N sources treatments were pooled, FA-induced increases in soil C content was most evident for FA5 (Fig. 3b). The increment of soil C by FA application relative to FA0 differed (P < 0.001) with N source (Table S2). Compared to FA0, FA-induced increase in soil C content was greatest in the vetch treatment (Fig. 7a). When C contained in the FA was excluded from the total soil C content, FA5 and FA10 increased soil C by 12.2 and 9.6%, respectively, in the CK, by 17.1 and 4.9%, respectively, in the urea, by 10.9 and 11.3%, respectively, in the compost, and by 28.0 and 30.3%, respectively, in the vetch treatment (Fig. 7b). These results show that the FA effects on soil C increment was the greatest when it was co-applied with vetch though the highest soil C content was recorded in the treatment with the co-application of FA and compost as reported above (Fig. 5). Among the different-sized particle fractions, the silt-sized fraction showed the largest (P < 0.001) increase in soil C content when compared to year 0; the C concentration in the silt-sized fraction was 3.6 g C kg−1 (representing 35.4% of the C in the bulk soil) in year 0 and it increased (range: 4.1–8.2 g C kg−1 which represents 44.0–54.5% of the C in the bulk soil) in year 3 (Table S2). The percentage increment of C content in the silt-sized fraction relative to the CK was greater for compost (25.0–46.4%) and vetch (4.9–33.9%) than for urea (1.8–9.8%) treatments regardless of FA application (Fig. 8a). When FA was coapplied, the greatest increment (36.8% for FA5 and 42.5% for FA10) of soil C in the silt-sized fraction relative to FA0 was observed in vetch treatment (Fig. 8b). The soil C loss during three rice growing seasons was affected (P < 0.001) by both N source and FA application rate (Table 3). Comparing the soil C loss among N sources at a given FA application rate, the greatest soil C loss was in vetch, while the lowest loss was in compost treatments. The application of FA consistently reduced (P < 0.001) soil C loss by 24.5–83.0% regardless of the N source and FA application rate.

Fig. 3. Changes in (a) total plant biomass (three years) and (b) soil C content (year 3 only) with fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). The values are the mean across the N sources, and vertical bars are standard errors of the mean.

rice seasons, the increment of grain yields relative to CK was in the order of vetch (41.0%) = urea (35.6%) > compost treatment (11.0%) (Fig. 2b and Fig. S2 for data of each year). However, grain yield was not affected (P = 0.167) by FA application rate (Table 2). The rice plant residue (straw and root) C that was incorporated into the soil was also affected by year (P < 0.001), N source (P < 0.001), and FA application rate (P = 0.031) in the same manner to that of total rice plant biomass (Table 2). On average across years, the increment of rice plant residue C relative to the CK was much greater in urea (54.0%) and vetch (43.9%) than in compost (5.7%) treatments (Figs. 2 c and S3 for data of each year). Application of FA reduced rice plant residue C by 5.9% for FA5 and by 11.0% for FA10 on average across the years (Fig. 2c). 3.2. Parameters of rice growth and grain yield All the rice growth and grain yield parameters varied (P < 0.001) with years in a similar pattern to that of total plant biomass (Table 2 and Figs. S4–6 for data of each year). The N source effects on increasing rice plant biomass and grain yield were achieved by increasing the number of tillers per m2 (P < 0.001) and the number of panicles per tiller (P < 0.001) rather than the number of grains per panicle (P = 0.10) or individual grain weight (P = 0.14) (Table 2). Specifically, both the number of tillers and panicles were greater (P < 0.001) in urea and vetch than in compost and CK treatments; i.e., across years, the number of tillers was 42.1 and 37.7% higher in urea and vetch, respectively, as compared to CK, whereas that of compost treatments did not differ from CK (Fig. 4). On the other hand, FA decreased (P = 0.047) the number of tillers per m2, but increased (P = 0.041) individual grain weight regardless of the N source and year (Table 2; Fig. 4).

4. Discussion 4.1. Effects of N source and FA rate on rice plant biomass and grain yield Among many weather factors affecting rice growth, in our study, higher temperature coupled with relatively greater solar radiation and rainfall in year 1 resulted in the greatest rice plant biomass; whereas, low rainfall in year 2 and two typhoons in year 3 during rice growing season (Figs. 1 and S1 for annual rice plant biomass) hampered rice growth. Annual variations in grain yield (Fig. S2), residue biomass (Fig. S3), and yield parameters such as individual grain weight (Figs. S4–6) also followed similar patterns to that of total rice plant biomass, as they were affected by weather conditions. In spite of the annual variations in rice growth, application of urea and vetch consistently increased rice

3.3. Soil C content and loss The soil C content was consistently affected (P < 0.001) by year, N 16

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Fig. 4. Three-year mean data of rice growth and yield parameters: (a) number of tillers per square meter, (b) number of panicles per tiller, (c) number of grains per panicle, and (d) individual grain weight of rice grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). Please see Fig. 2 for the detailed description of the figure. Annual data are provided in Supporting information (Figs. S4–6).

biomass and yield over compost application via increased tiller number and panicle number due to high N availability in the soils treated with urea and vetch (Figs. 2 and 4) as supported by Asagi and Ueno (2009) and Yun et al. (2011). The negative impact of FA application on total rice plant biomass for FA10 (Figs. 2 a and 3 a) or the insignificant effect on grain yield for both

FA5 and FA10 (Fig. 2c) contradict with many studies that reported FA amending at < 12% (w/w) improved rice growth and yield by ameliorating soil pH and supplying nutrients (Lee et al., 2003, 2006; Dwivedi et al., 2007; Singh et al., 2012). Therefore, this result suggests that the FA amendment does not always improve rice growth as also reported by Lim et al. (2016). In this study, even though the FA application rates 17

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Fig. 5. Annual changes in soil C content across three rice cropping seasons. Rice was grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). Data were provided for both when FA-derived C was included (a–d) and FA-derived C was excluded (e–h). Vertical bars are standard errors of the mean (error bars are often too small to be visible). Detailed treatments are provided in Table S1. The effect of year (Y), N source (N), FA application rate (FA), and their interaction on soil C content were all significant (P < 0.01).

were lower than the critical rate (i.e., 12%), rice growth might be hampered by adverse impacts of FA such as high pH, salinity, and substantial concentrations of toxic elements such as B and As in FA and

thus the benefits of FA amendment could be masked (Basu et al., 2009; Lim et al., 2012a). In addition, P immobilization by Ca contained in FA could also deteriorate rice growth by decreasing P availability (Singh

Fig. 6. Changes in soil C content at the end of the three-year rice cropping seasons relative to the initial. Data were provided for both when FA-derived C was included (a) and FA-derived C was excluded (b). Rice was grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). No error bar is provided as the percentage changes were calculated using the mean value of each treatment. Detailed treatments are in Table S1.

18

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Fig. 7. Changes in soil C content of flay ash (FA)-amended soils relative to FA-free soil at the end of the three-year rice cropping seasons. Data were provided for both when FA-derived C was included (a) and FA-derived C was excluded (b). Rice was grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). No error bar is provided as the percentage changes were calculated using the mean value of each treatment. Detailed treatments are provided in Table S1.

decomposition not by increasing plant biomass. These results highlight the critical roles of humified organic amendment such as livestock manure compost, rather than the amount of rice plant residues incorporated into the soils, in maintaining or increasing the soil C level (Lee et al., 2009). Due to the higher chemical stability of livestock manure compost than green manure, more C can be retained in the soils when compost rather than green manure was applied (Nyberg et al., 2002; Lim et al., 2012b). Rice plant residue incorporation is reported to be a more effective measure to increase soil C content than other BMPs including no tillage or balanced fertilization over the corresponding control treatments (e.g., straw removal, conventional tillage, and imbalanced fertilization) (Liu et al., 2014; Tian et al., 2015; Zhu et al., 2015). Many studies in rice cultivating countries including China (Ji et al., 2011; Zhu et al., 2015 and references cited therein), India (Mandal et al., 2007), Japan (Minamikawa and Sakai, 2007), and Korea (Lee et al., 2009) also reported that rice straw incorporation increased soil C over the control from which straw was removed. However, it should be also noticed that soil C still declined or increased marginally with rice cropping seasons even with straw incorporation, and thus crop residue incorporation merely slows down soil C depletion (Fig. 5). This could be ascribed to stimulated decomposition of native soil organic C (Fontaine et al., 2004; Baek et al., 2011; Ye et al., 2015) via the so-called priming effect, an acceleration of decomposition of native soil organic C by the supply of fresh organic C that stimulates microbial activities (Kuzyakov et al., 2000), as well as the rapid decomposition (>80% decomposition in a single cultivation season) of rice plant residues (Viswanath et al., 2010). Our results also demonstrate that rice straw incorporation had minimal effects on soil C content over cropping seasons due to the readily decomposable nature of rice plant residues. The increases in soil C by FA application could be primarily due to the input of black C contained in FA that is recalcitrant to microbial decomposition (Koschke et al., 2011). However, the increase in soil C in

et al., 2012; Lim et al., 2016). As the decreased rice plant biomass including straw and root by FA application (particularly for FA10) was driven by reduced number of tillers (Table 2; Fig. 4), we propose that lowered P availability by FA application together with other constraints above-mentioned hampered tillering at the early growing stage. Nutrient availability and rhizosphere environmental conditions at the early growing state play decisive roles in tiller formation (Kim et al., 2003; Fageria, 2014) as also evidenced by our previous study (Lim et al., 2016). Interestingly, though FA application reduced total rice plant biomass by decreasing the number of tillers, the grain yield was not affected by FA application (Table 2, Fig. 2) due to the increase in individual grain weight with FA application (Table 2, Fig. 4d) that redeemed the grain yield loss. Individual grain weight is known to be a major contributor in increasing grain yield for high-yielding rice cultivars (Yoshida, 1981). Increases in individual grain weight coupled with the reduced number of tiller or panicle (Table 2 and Fig. 4) is often reported to be due to increased allocation of photosynthates to an individual grain particularly when nutrients such as N (Li et al., 2016) and P (Fageria, 2014) are limited that hampers the formation of tillers and panicles. Therefore, our results suggest that FA application may not affect rice grain yield, but may decrease the soil C sequestration potential in paddy rice soils given that the amount of plant residue incorporated into soils is critical in increasing soil C content (Lal, 2004; Liu et al., 2014). 4.2. Effects of N source and FA rate on soil C content and loss Among the N sources applied, in the absence of FA, compost was the most efficient in increasing soil C content (Fig. 6a) despite that compost treatment produced less rice plant residues than urea and vetch treatments (Fig. 2c). These findings suggest that compost increases soil C content via supplying organic C that is recalcitrant to microbial

Fig. 8. Changes in C content of silt-sized particles by N source and fly ash (FA) application at the end of the three-year rice cropping seasons. Rice was grown with different N sources (CK, no input; urea; compost, pig manure compost; vetch, hairy vetch green manure) and fly ash (FA) application rates (FA0, 0%; FA5, 5%; FA10, 10% by dry weight of soil). The changes by N source and FA application rate were presented relative to CK at a given FA application rate (a) and relative to FA0 at a given N source (b), respectively. No error bars are provided because the percentage changes were calculated with the mean values of each treatment. Detailed treatments are provided in Table S1.

19

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Table 3 Soil (0–20 cm) carbon (SC) balance in rice paddy field during three rice cultivation seasons as affected by N source (no input, urea, pig manure compost (compost), and hairy vetch green manure (vetch)) and fly ash (FA) application rate (FA0, 0%; FA5, 5%; FA10, 10% by soil weight). N source

FA application rate

SC in Year 0 (g C m−2) (A)

C input (g C m−2)

SC in Year 3a (g C m−2) (E)

FA (B)

N source (C)

Residue (D)a

SC loss (A + B + C + D−E)b Amount (g C m−2)a

Relative to FA0 (%)

No input

FA0 FA5 FA10

2658 2658 2658

0 260 520

0 0 0

535 (24)bc 460 (30)cd 423 (10)d

2289 (206)f 2828 (83)de 3030 (70)cd

904 (206)b 550 (72)de 571 (60)de

0 −39.1 −36.9

Urea

FA0 FA5 FA10

2658 2658 2658

0 260 520

14 14 14

706 (24)a 626 (33)ab 637 (19)a

2579 (133)ef 3280 (18)c 3225 (63)c

799 (133)c 278 (50)f 603 (46)cd

0 −65.3 −24.5

Compost

FA0 FA5 FA10

2658 2658 2658

0 260 520

354 354 354

498 (43)cd 522 (45)cd 449 (23)cd

3037 (82)cd 3629 (10)b 3900 (114)a

473 (83)e 165 (54)g 80 (134)g

0 −65.1 −83.0

Vetch

FA0 FA5 FA10

2658 2658 2658

0 260 520

379 379 379

665 (27)a 662 (54)a 625 (21)ab

2399 (123)f 3331 (101)c 3646 (37)b

1302 (123)a 628 (152)cd 536 (54)de

0 −51.8 −58.8

P values < 0.001 0.021 0.571

< 0.001 < 0.001 0.094

< 0.001 < 0.001 0.150

ANOVA results N source (N) Fly ash application rate (FA) N × FA a

Values are means with the standard errors of triplicate measurement in parentheses, and those with the same letters are not significantly different (P > 0.05) among the treatments. Crop residue C after the harvest in year 3 was not included in the SC balance analysis as soil C content was measured at the harvest in year 3 before residue incorporation. The value of “SC loss” indicates the amount of soil C lost during the 3-yrs experiment, and the negative value of “Relative to FA0” indicates that C loss was reduced by FA application. b

compatible with the traditional BMP (e.g., organic amendment) for enhanced soil C sequestration.

the FA-amended soils even after excluding the C derived from FA (Figs. 3b and 7b) proposes additional mechanisms of FA-induced increases in soil C since the quantities of rice plant residues C incorporated into the soils were reduced by FA application (Fig. 2c). As soil C content is determined by the balance between C input and C loss (Lal, 2004), such increases in soil C by FA in spite of lower residue-C incorporation suggests that FA increased soil C content by reducing C loss (Table 3). Suppression of microbial decomposition of soil C and chemical fixation of CO2 in the form of carbonate precipitates were proposed as mechanisms for increasing soil C by FA application (Lim et al., 2012a; Lim and Choi, 2014). In addition, increased pH (range: 6.53–7.49, Table S3) by FA application favors the reaction of Ca2+ with CO32− to form calcium carbonates (Azdarpour et al., 2015), which is considerably more stable compared to organic C (Lackner, 2003). In addition, this study further suggests that the higher silt-sized fraction in FA might have also contributed to reducing the soil C loss via physical protection (Six et al., 2004; Jastrow et al., 2007) as indicated by the greater increment of soil C concentration in the silt-sized fraction than in other fractions (Table S2 and Fig. 8). Yunusa et al. (2015) reported that the ability of FA to adsorb organic C was low in a batch experiment. However, it was reported that FA application substantially decreased total organic C concentration in paddy water during the rice growing season in an experiment with the same FA used as in this study (Ham, 2015), indicating that FA can reduce C loss through sorption of dissolved organic C onto FA particles. Such FA-induced reduction in soil C loss via microbiological, chemical, and physical mechanisms was most apparent when vetch was co-applied with FA (Fig. 7). This result is supported by Lim and Choi (2014) that reported that the CO2 flushed from rapid decomposition of vetch can be retained in the soils by carbonation of Ca or Mg contained in FA and thus magnifying the FA effect on soil C content. Therefore, our study together with other publications (Lim et al., 2012a; Lim and Choi, 2014; Ham, 2015) suggest that microbiological (suppression of microbial growth), chemical (carbonation), and physical (protection in the silt-sized fraction) pathways are responsible for the reduced soil C loss and thus increasing soil C by FA application. Such effect of FA on decreasing soil C loss and thus increasing soil C content was more consistent across FA application rate when organic fertilizers including compost and vetch rather than urea were co-applied with FA. This result suggests that FA application is

5. Conclusions Our study shows that FA is a promising soil amendment that enhances soil C sequestration by reducing soil C loss without hampering rice grain yield regardless of the N source. As increased FA application rate further reduced soil C loss when FA was co-applied with compost and vetch, FA may be more efficient in reducing soil C loss when organic N sources rather than synthetic N fertilizer are used. The FAinduced soil C sequestration was most evident when vetch was applied, and the rice grain yield in the vetch treatment was comparable to the urea treatment. Therefore, to achieve dual goals of sustainable rice grain yield and enhanced soil C sequestration, co-application of FA with green manure would be a more feasible strategy than with synthetic N fertilizer or livestock manure compost. We conclude that mineral amendments that have ability to reduce soil C loss can be an alternative BMP for enhanced soil C sequestration.

Acknowledgements This work was supported by the Cooperative Research Program for Agricultural Science & Technology Development (Project No. PJ007409032011) of Rural Development Administration and by the National Research Foundation funded by the Ministry of Education (NRF-2015R1D1A3A01018961), Republic of Korea. We thank to Kwang-Seung Lee, Se-In Lee, Byung-Jun Jeon, Jong-Hyun Ham, Hyun-Jin Park, Won-Sung Kim, Geun-Bae Kim, Seo-Jin You, So-Dam Choi, and Jung-Yun Um for their assistance with field works and laboratory sample analyses. The authors declare no conflict of interest.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.agee.2017.03.027. 20

Agriculture, Ecosystems and Environment 245 (2017) 11–21

S.-S. Lim et al.

Carbon mineralization and retention of livestock manure composts with different substrate qualities in three soils. J. Soils Sediments 12, 312–322. Lim, S.S., Lee, D.S., Kwak, J.H., Park, H.J., Kim, H.Y., Choi, W.J., 2016. Fly ash and zeolite amendments increase soil nutrient retention but decrease paddy rice growth in a low fertility soil. J. Soils Sediments 16, 756–766. Liu, C., Lu, M., Cui, J., Li, B., Fang, C.M., 2014. Effects of straw carbon input on carbon dynamics in agricultural soils: a meta-analysis. Glob. Change Biol. 20, 1366–1381. Mandal, B., Majumder, B., Bandyopadhyay, P.K., Hazra, G.C., Gangopadhyay, A., Samantaray, R.N., Mishra, A.K., Chaudhury, J., Saha, M.N., Kundu, S., 2007. The potential of cropping systems and soil amendments for carbon sequestration in soils under long-term experiments in subtropical India. Glob. Change Biol. 13, 357–369. Minamikawa, K., Sakai, N., 2007. Soil carbon budget in a single-cropping paddy field with rice straw application and water management based on soil redox potential. Soil Sci. Plant Nutr. 53, 657–667. Nayak, A.K., Raja, R., Rao, K.S., Shukla, A.K., Mohanty, S., Shahid, M., Tripathi, R., Panda, B.B., Bhattacharyya, P., Kumar, A., Lal, B., Sethi, S.K., Puri, C., Nayak, D., Swain, C.K., 2015. Effect of fly ash application on soil microbial response and heavy metal accumulation in soil and rice plant. Ecotox. Environ. Safe. 114, 257–262. Nyberg, G., Ekblad, A., Buresh, R., Högberg, P., 2002. Short-term patterns of carbon and nitrogen mineralisation in a fallow field amended with green manures from agroforestry trees. Biol. Fertil. Soils 36, 18–25. Pandey, V.C., Singh, N., 2010. Impact of fly ash incorporation in soil systems. Agric. Ecosyst. Environ. 136, 16–27. Park, H.J., Lim, S.S., Kwak, J.H., Baek, W.J., Yoon, K.S., Choi, S.M., Choi, W.J., 2015. Nitrogen inputs with different substrate quality modified pH, Eh, and N dynamics of a paddy soil incubated under waterlogged conditions. Commun. Soil Sci. Plant Anal. 46, 2234–2248. Rui, W.Y., Zhang, W.J., 2010. Effect size and duration of recommended management practices on carbon sequestration in paddy field in Yangtze delta plain of China: a meta-analysis. Agric. Ecosyst. Environ. 135, 199–205. Rural Development Administration, 2000. Taxonomical Classification of Korean Soils. RDA of Korea, Suwon Republic of Korea. Sahrawat, K.L., 2004. Organic matter accumulation in submerged soils. Adv. Agron. 81, 169–201. Singh, A., Sarkar, A., Agrawal, S.B., 2012. Assessing the potential impact of fly ash amendments on Indian paddy field with special emphasis on growth, yield, and grain quality of three rice cultivars. Environ. Monit. Assess. 184, 4799–4814. Six, J., Bossuyt, H., Degryze, S., Denef, K., 2004. A history of research on the link between (micro) aggregates, soil biota, and soil organic matter dynamics. Soil Tillage Res. 79, 7–31. Tian, K., Zhao, Y.C., Xu, X.H., Hai, N., Huang, B.A., Deng, W.J., 2015. Effects of long-term fertilization and residue management on soil organic carbon changes in paddy soils of China: a meta-analysis. Agric. Ecosyst. Environ. 204, 40–50. Viswanath, T., Pal, D., Purakayastha, T.J., 2010. Elevated CO2 reduces rate of decomposition of rice and wheat residues in soil. Agric. Ecosyst. Environ. 139, 557–564. West, T.O., Six, J., 2007. Considering the influence of sequestration duration and carbon saturation on estimates of soil carbon capacity. Clim. Change 80, 25–41. Yan, H.M., Cao, M.K., Liu, J.Y., Tao, B., 2007. Potential and sustainability for carbon sequestration with improved soil management in agricultural soils of China. Agric. Ecosyst. Environ. 121, 325–335. Ye, R.Z., Doane, T.A., Morris, J., Horwath, W.R., 2015. The effect of rice straw on the priming of soil organic matter and methane production in peat soils. Soil Biol. Biochem. 81, 98–107. Yoshida, S., 1981. Fundamentals of Rice Crop Science. International Rice Research Institute, Los Baños, Philippines. Yun, S.I., Lim, S.S., Lee, G.S., Lee, S.M., Kim, H.Y., Ro, H.M., Choi, W.J., 2011. Natural 15 N abundance of paddy rice (Oryza sativa L.) grown with synthetic fertilizer livestock manure compost, and hairy vetch. Biol. Fertil. Soils 47, 607–617. Yunusa, I.A.M., Blair, G., Zerihun, A., Yang, S.J., Wilson, S.C., Young, I.M., 2015. Enhancing carbon sequestration in soil with coal combustion products: a technology for minimising carbon footprints in coal-power generation and agriculture. Clim. Change 131, 559–573. Zhu, L.Q., Li, J., Tao, B.R., Hu, N.J., 2015. Effect of different fertilization modes on soil organic carbon sequestration in paddy fields in South China: a meta-analysis. Ecol. Indic. 53, 144–153.

References Asagi, N., Ueno, H., 2009. Nitrogen dynamics in paddy soil applied with varioius 15Nlabelld green manures. Plant Soil 322, 251–262. Azdarpour, A., Asadullah, M., Mohammadian, E., Hamidi, H., Junin, R., Karaei, M.A., 2015. A review on carbon dioxide mineral carbonation through pH-swing process. Chem. Eng. J. 279, 615–630. Baek, W.J., Kim, Y.J., Yun, S.I., Lee, S.I., Lim, S.S., Kim, H.Y., Yoon, K.S., Choi, S.M., Choi, W.J., 2011. Sequestration of roots-derived carbon in paddy soil under elevated CO2 with two temperature regimes as assessed by isotope technique. J. Korean Soc. Appl. Biol. Chem. 54, 403–408. Basu, M., Pande, M., Bhadoria, P.B.S., Mahapatra, S.C., 2009. Potential fly-ash utilization in agriculture: a global review. Prog. Nat. Sci. 19, 1173–1186. Blissett, R.S., Rowson, N.A., 2012. A review of the multi-component utilization of coal fly ash. Fuel 97, 1–23. Dwivedi, S., Tripathi, R.D., Srivastava, S., Mishra, S., Shukla, M.K., Tiwari, K.K., Singh, U.N., Rai, U.N., 2007. Growth performance and biochemical responses of three rice (Oryza sativa L.) cultivars grown in fly-ash amended soil. Chemosphere 67, 140–151. Fageria, N.K., 2014. Yield and yield components and phosphorus use efficiency of lowland rice genotypes. J. Plant Nutr. 37, 979–989. Fontaine, S., Bardoux, G., Abbadie, L., Mariotti, A., 2004. Carbon inputs to soil may decrease soil carbon content. Ecol. Lett. 7, 314–320. Ham, J.H., 2015. Changes in Ponding Water Quality and Rice Growth in Paddy Fertilized with Liquid Pig Manure as Affected by Fly Ash and Zeolite Amendments. MSc Thesis. Chonnam National University, Gwangju, Republic of Korea. Jala, S., Goyal, D., 2006. Fly ash as a soil ameliorant for improving crop production – a review. Bioresour. Technol. 97, 1136–1147. Jastrow, J.D., Amonette, J.E., Bailey, V.L., 2007. Mechanisms controlling soil carbon turnover and their potential application for enhancing carbon sequestration. Clim. Change 80, 5–23. Ji, X.H., Wu, J.M., Peng, H., Shi, L.H., Zhang, Z.H., Liu, Z.B., Tian, F.X., Huo, L.J., Zhu, J., 2011. The effect of rice straw incorporation into paddy soil on carbon sequestration and emissions in the double cropping rice system. J. Sci. Food Agric. 92, 1038–1045. Kim, H.Y., Lieffering, M., Kobayashi, K., Okada, M., Miura, S., 2003. Seasonal changes in the effects of elevated CO2 on rice at three levels of nitrogen supply: a free air CO2 enrichment (FACE) experiment. Glob. Change Biol. 9, 826–837. Koschke, L., Lorz, C., Fürst, C., Glaser, B., Makeschin, F., 2011. Black carbon in fly-ash influenced soils of Dübener Heide region, central Germany. Water Air Soil Pollut. 214, 119–132. Kuzyakov, Y., Friedel, J.K., Stahr, K., 2000. Review of mechanisms and quantification of priming effects. Soil Biol. Biochem. 32, 1485–1498. López, M., Huerta-Pujol, O., Martínez-Farré, F.X., Soliva, M., 2010. Approaching compost stability from Klason lignin modified method: chemical stability degree for OM and N quality assessment. Resour. Conserv. Recycl. 55, 171–181. Lackner, K.S., 2003. A guide to CO2 sequestration. Science 300, 1677–1678. Lal, R., 2004. Soil carbon sequestration to mitigate climate change. Geoderma 123, 1–22. Lee, Y.B., Ha, H.S., Lee, K.D., Park, K.D., Cho, Y.S., Kim, P.J., 2003. Evaluation of use of fly ash-gypsum mixture for rice production at different nitrogen rates. Soil Sci. Plant Nutr. 49, 69–76. Lee, H., Ha, H.S., Lee, C.H., Lee, Y.B., Kim, P.J., 2006. Fly ash effect on improving soil properties and rice productivity in Korean paddy soils. Bioresour. Technol. 97, 1490–1497. Lee, S.B., Lee, C.H., Jung, K.Y., Park, K.D., Lee, D.K., Kim, P.J., 2009. Changes of soil organic carbon and its fractions in relation to soil physical properties in a long-term fertilized paddy. Soil Tillage Res. 104, 227–232. Lee, S.I., Lim, S.S., Lee, K.S., Park, W.K., Shin, J.D., Yoon, K.S., Kim, H.Y., Choi, W.J., 2014. Coal fly ash enhanced planted-floating bed performance in phosphoruscontaminated water treatment. Ecol. Eng. 73, 276–280. Li, G.H., Zhang, J., Yang, C.D., Song, Y.P., Zheng, C.Y., Liu, Z.H., Wang, S.H., Tang, S., Ding, Y.F., 2016. Yield and yield components of hybrid rice as influenced by nitrogen fertilization at different eco-sites. J. Plant Nutr. 37, 244–258. Lim, S.S., Choi, W.J., 2014. Changes in microbial biomass CH4 and CO2 emissions, and soil carbon content by fly ash co-applied with organic inputs with contrasting substrate quality under changing water regimes. Soil Biol. Biochem. 68, 494–502. Lim, S.S., Choi, W.J., Lee, K.S., Ro, H.M., 2012a. Reduction in CO2 emission from normal and saline soils amended with coal fly ash. J. Soils Sediments 12, 1299–1308. Lim, S.S., Lee, K.S., Lee, S.I., Lee, D.S., Kwak, J.H., Hao, X.Y., Ro, H.M., Choi, W.J., 2012b.

21