Published in Agron. J. 96:70-76 (2004).
© American Society of Agronomy
677 S. Segoe Rd., Madison, WI 53711 USA
RICE
Impact of Rice Straw Incorporation on Soil Redox Status and Sulfide Toxicity
Suduan Gao*,
Kenneth K. Tanji and
Steven C. Scardaci
Dep. of Agron. and Range Sci., Univ. of California, Davis, CA 95616
* Corresponding author (sugao{at}ucdavis.edu).
Received for publication February 25, 2002.
 |
ABSTRACT
|
|---|
Incorporation of rice (Oryza sativa L.) straw into the soil has become an alternative to straw burning to improve air quality in the Sacramento Valley, California. However, straw addition to paddies could promote reducing conditions that may lead to sulfide accumulation and plant toxicity. Sulfide toxicity has been observed in randomly localized field sites and is characterized by blackened roots, retarded growth, fewer standing plants, and even death in severe cases. The objective of this study was to investigate the impact of straw incorporation on soil redox status and sulfide toxicity to rice in a greenhouse pot study. Treatments included straw incorporation (0, 6, and 23 g straw kg–1 soil) and sulfate additions (0, 160, and 800 mg SO4 kg–1 soil). Redox status was evaluated by identifying dominant terminal electron-accepting processes and geochemical redox classes based on oxidative capacity. Higher straw incorporation rates led to more reducing conditions at earlier times. The most reducing conditions (methanic) were observed within 3 wk for the 23 g straw kg–1 soil treatment and in about 6 wk for the 6 g straw kg–1 soil treatment and were not observed till the end of the experiment when no straw was added. Straw incorporation significantly reduced grain yield (p < 0.0001), number of tillers (p < 0.0001), and plant height at 4 wk (p = 0.01). Sulfate addition only showed significant reduction on the number of tillers (p = 0.0028). Soluble sulfide concentrations were very low, mainly due to precipitation with Fe. The higher straw incorporation rates induced sulfide toxicity symptoms and reduced rice yield significantly. It is not clear, however, if other causes, such as organic acids and salinity, may have contributed to the adverse impact of straw incorporation on rice.
Abbreviations: DO, dissolved oxygen • EC, electrical conductivity • EH, redox potential • OXC, oxidative capacity • SI, saturation index
|
INTRODUCTION
|
|---|
PADDY RICE is one of the major crops grown in the Sacramento Valley of California. The total annual planting of rice fluctuates from 160000 to 240000 ha, depending on market and weather conditions. Traditional rice straw burning after harvest has been restricted due to its contribution to poor air quality and subsequent California air quality legislation (the Connelly–Areia–Chandler Rice Straw Burning Reduction Act of 1991 and Senate Bill 318 in 1999). Since few commercial uses of rice straw have been found, rice growers have chosen to incorporate straw into the soil as a major alternative to burning. The average straw return rate is about 4 g straw kg–1 soil. Straw incorporation requires additional management practices such as rolling and flooding to enhance straw decomposition and increases the incidence of weeds and crop diseases. In addition, straw incorporation could promote reducing conditions under which toxic products such as sulfide may be produced and cause toxicity to rice plants.
When a paddy soil is submerged, the diffusion of gases into the soil is drastically reduced. Aerobic microorganisms during mineralization of soil organic matter can quickly deplete dissolved oxygen (DO), and the soil becomes anaerobic (Ponnamperuma, 1981). Subsequently, decomposition of organic materials will be dominated by facultative or obligate anaerobic microorganisms. While oxidizing organic matter, these organisms use inorganic components such as nitrate, Mn oxyhydroxides, Fe oxyhydroxides, sulfate, and CO2 or reducible organic compounds (fermentation) as electron acceptors in completing the redox reactions (Patrick, 1981). Although there is a theoretical sequence in reducing these components, i.e., O2, NO–3, Mn(III, IV), Fe(III), SO2–4, and CO2 after flooding, significant overlapping of using these electron acceptors has been observed in the rice field near Maxwell, CA (Gao et al., 2002). These redox processes yield products such as fatty acids and hydrogen sulfide that may affect rice plants.
Field studies near Maxwell in the Sacramento Valley have shown that straw incorporation did enhance reducing conditions (Gao et al., 2002). Observations in the field plots revealed that straw incorporation into the soil was not likely to cause an adverse effect on rice plants on a large scale. However, sulfide toxicity symptoms on rice in randomly localized sites were observed mostly in the drain outlet ends of the plots. Sulfide toxicity symptoms were observed with blackened roots, much shorter plant height, and fewer number of standing plants in addition to the rotten-egg odor of collected soil samples in the affected area compared with the healthy plant areas. The black color on the roots disappeared after a few hours of exposure to the atmosphere due to reoxidation of sulfide. Other literature-reported symptoms for sulfide toxicity on rice included gray-green colored leaves, sterile florets and subsequent reduced grain yield, and death in severe cases (Kuo and Mikkelsen, 1981; Allam and Hollis, 1972).
There are some difficulties to diagnose sulfide toxicity by testing the soil due to low solution sulfide concentrations. The free hydrogen sulfide is the toxic sulfide form to rice, but its solubility is constrained by precipitation with metal ions, especially Fe(II) (Yoshida, 1981). Sulfide concentrations from 0.07 to 0.1 mg L–1 significantly inhibited the respiration of rice roots (Allam and Hollis, 1972). It is known that precipitation of amorphous or other forms of FeS has direct impact on soluble sulfide concentrations (Yoshida, 1981; Neue and Bloom, 1989). Chemical speciation programs can be utilized to evaluate sulfide species (H2S, HS–, and S2–) and precipitation of sulfide with Fe(II). Our hypothesis is that straw incorporation enhances reducing conditions and sulfate is reduced to sulfide or H2S gas. Sulfide formation, solubility, and toxicity are controlled by the dynamic soil conditions and thus affected by factors such as straw incorporation level, source of sulfate, and reductions of soil Mn and Fe. In the presence of soluble Fe(II) and Mn(II), formed from the reduction of Fe(III) and Mn (IV, III) minerals, sulfide solubility is regulated by the formation of Fe and Mn sulfides. However, when electron acceptors of Mn (IV, III) and Fe(III) oxyhydroxides become depleted, sulfide may accumulate in the soil solution. Thus, the occurrence of sulfide toxicity in paddy rice depends on a host of specific conditions in the soil. We designed a greenhouse study with the objective to investigate the impact of straw incorporation and sulfate addition on redox status and subsequent sulfide toxicity to rice. The results may be valuable in interpreting the conditions that favor sulfide formation and toxicity to rice plants and defining the potential of sulfide toxicity to rice due to straw incorporation.
|
MATERIALS AND METHODS
|
|---|
Soil
The soil was collected from a field where sulfide toxicity was observed in a rice straw management project near Maxwell, CA. Surface soil was collected near the drain outlet area in a straw-burned plot. The soil is Willows clay (fine, smectitic, thermic Sodic Endoaquert) having an electrical conductivity (EC) of 2.5 dS/m and pH of 7.4 in a 1:1 soil/water suspension.
Pot Study and Treatment
Soil treatments involved three levels of straw, three levels of sulfate, and their interactions (Table 1). Three replicates were used in this study.
Soil taken from surface was mixed thoroughly. Then nine equal amounts of the soil, each with about 13 kg of air-dry soil, were weighed. The rice straw was added at rates of 0, 6, and 23 g straw kg–1 soil and is denoted as St0, St1, and St2 treatments, respectively. The straw used in this study was obtained from the Maxwell experimental site after harvest in 1999, dried, and then cut into 2 to 3 cm length. After mixing the straw thoroughly with the soil, the soil was then placed into PVC pots that had an inside diameter of 20 cm and a height of 20 cm.
To each pot, (containing 4.3 kg air-dry soil), SO4 as (NH4)2SO4 was added at rates of 0, 160, and 800 mg SO4 kg–1 soil and is denoted by the S0, S1, and S2 treatments, respectively. For fertilization, a total of 1 g of N either as urea or (NH4)2SO4 or both was applied to each pot to balance the sulfate treatment level. The S0 treatment received 2.145 g of urea. The S1 treatment received 1.716 g of urea and 0.944 g of (NH4)2SO4. The S2 treatments received 4.719 g of (NH4)2SO4. In addition, about 0.12 g of K and 0.07 g of P was applied by adding 0.167 g of KH2PO4 to each pot. All of the chemicals were dissolved in about 3 L of distilled water and added to the pots at initial flooding on 27 June 2000.
The soil levels in each pot were kept about 5 cm below the top edge of the pot. Water levels were kept close to the top of the pot during most of the growing season; except at the beginning, the water levels were 2 to 3 cm deep. Ten seeds of rice (M202) in germination were planted in each pot on 28 June 2000.
Pore waters were sampled for the first time 24 to 48 h after flooding and then after about 1-, 2-, and 3-wk intervals. Sampling ports were installed at about 2 cm from the bottom of the PVC pots. A vacuum pumping technique was used for on-line monitoring of DO and redox potential (EH). Soil solution sulfide concentration was determined by an ion selective electrode (Model 9416, Orion Res., Beverly, MA). Other solution parameters acquired included pH, EC, NO3, Fe(II) and total Fe(II + III), dissolved Mn(II), and sulfate concentrations. Dissolved gas concentrations such as methane and hydrogen gas (H2) were also determined. Soil redox status was evaluated by deriving dominant terminal electron-accepting processes (TEAPs) and classifications based on oxidative capacity (OXC) calculations. Detailed information on sample collection and analysis and application of redox status indicators can be found in Gao et al. (2002)
Statistical Analysis
The treatment effects on number of tillers, plant height at early growing stages, and grain yield in the greenhouse study were performed using PC/SAS version 8.01 (SAS Inst., 1991). Comparisons among straw incorporation, sulfate addition, and their interaction are analyzed.
|
RESULTS AND DISCUSSION
|
|---|
Redox Potential of Soil Solution
Changes in EH of soil solution for each treatment are shown in Fig. 1. The pH effect on EH values is expectedly very small as initial soil solution pH (about 7.5) rapidly approached neutral on flooding and was maintained neutral throughout the growing season. The changes in EH strongly reflected the levels of straw added to the soils, but large variability in EH was recorded in the first few weeks after flooding. Higher straw addition resulted in lower EH and faster rates of EH decrease. After 1 d of flooding, soil pore water EH for the 23 g straw kg–1 soil treatment was approximately 100 mV or lower and, one month later, reached the lowest level and was almost stable thereafter. The highest EH was observed in the no-straw addition treatment that gradually decreased and finally reached the lower levels after about 8 wk of flooding. Soil in the 6 g straw kg–1 soil treatment exhibited intermediate EH values. After 8 wk of flooding, differences in EH values among the treatments were small.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 1. Redox potential (EH) in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
Terminal Electron-Accepting Processes
Reduction of oxygen, nitrate, Mn oxyhydroxides, Fe oxyhydroxides, sulfate, and methane production were determined by monitoring some of the reactant and product concentrations. Figures 2 through 8 show changes in concentrations of DO, NO3––N, Mn(II), Fe(II), SO2–4–S, S2––S and methane in pore waters from initial flooding to drainage before harvest.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Dissolved oxygen (DO) concentrations in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 8. Dissolved methane concentration in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 3. Nitrate concentrations in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 4. Manganese (II) concentrations in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 5. Ferrous Fe concentrations in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 6. Sulfate concentration in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 7. Sulfide concentration (H2S and HS–) in soil solution. (Treatments: St0 = 0 g straw kg–1 soil; St1 = 6 g straw kg–1 soil; St2 = 23 g straw kg–1 soil; S0 = 0 mg SO4 kg–1 soil; S1 = 160 mg SO4 kg–1 soil; S2 = 800 mg SO4 kg–1 soil.)
|
|
Oxygen and nitrate concentration decreased rapidly on initial flooding, indicating reduction occurred (Fig. 2 and 3). In the 23 g straw kg–1 soil treatment, both DO and NO–3–N concentrations dropped to below detection levels within 24 to 48 h, and in other treatments, DO ranged from 0.11 to 0.17 mmol L–1 and NO–3–N from 0.34 to 0.46 mmol L–1.
Increases in the concentrations of Mn(II) and Fe(II) (Fig. 4 and 5) indicate reduction of Mn and Fe oxyhydroxides occurred next. After reaching maximum concentrations, the decrease in metal concentrations is considered to reflect formation of precipitates such as with sulfide, produced from sulfate reduction, and possibly carbonates too.
The maximum concentrations of Mn(II) were observed 3 wk after flooding with straw incorporation and 6 wk after flooding when no straw was added, indicating Mn reduction started and finished within 3 to 6 wk after flooding.
Iron reduction showed the same trend but was more strongly affected by the straw treatment (Fig. 5). The maximum concentration of Fe(II) was the highest (1.4 mmol L–1) with the 23 g straw kg–1 soil treatment and appeared earlier around 3 wk after flooding compared with 6 g straw kg–1 soil and no-straw treatments. The maximum concentration of Fe(II) for 6 g straw kg–1 soil treatment was much lower (0.6 mmol L–1) and appeared around 6 wk after flooding. With no straw added, Fe(II) concentration gradually increased until the end of the study where it approached a maximum concentration of 0.4 mmol L–1. Interestingly, the final concentrations of Fe(II) for all the treatments were similar. The results strongly indicate that straw addition directly affected the kinetics of Fe oxyhydroxide reduction. Iron served as an electron acceptor, and apparently the source was depleted within a short time after flooding for the highest level of straw treatment but continued to serve as an electron acceptor in the no-straw treatment. This soil contained a substantial amount of Fe that may play an important role in reducing sulfide toxicity by forming FeS. On the other hand, higher amount of straw incorporation could deplete the Fe source within a relatively short time and thus may stimulate sulfide toxicity from sulfate reduction.
Sulfate reduction also began to occur very early, immediately after flooding as observed by the exponentially decreasing sulfate concentration (Fig. 6). Analysis of variance and covariance showed that there was a positive correlation between initial measured sulfate concentration and sulfate added (r2 = 0.63, p value <0.0001). Initially measured sulfate concentrations were negatively correlated to straw treatment (r2 = 0.64, p value <0.0001). It is likely that sulfate reduction occurred within the first couple of days in the straw treatment that resulted in lower sulfate concentrations in the first sampling that was taken after 24 to 48 h of flooding.
The solution sulfide concentration was monitored (Fig. 7), but it cannot quantitatively represent sulfate reduction as can some of the sulfide-formed Fe precipitates. Sulfide concentrations for all samples were in a low range. Most of the samples contained sulfide concentrations below 0.005 mmol L–1 with two exceptionally high values with large deviations in the late season in the treatments where sulfate was added but not straw whereas the adverse effect of straw treatment on rice plants was observed at the early growing stages. Interpretation of the sulfide results is discussed focusing more on its solubility control.
Speciation model WATEQ (Ball et al., 1987) was used to test the formation of FeS by computing the saturation index (SI) of FeS. The SIs were calculated from the ion activity product (IAP) and the equilibrium constant (Keq), i.e., SI = log[IAP/Keq]. A positive SI indicates that the solution is likely to be oversaturated with respect to the solid phase mineral, or in other words, precipitation may occur. For the treatment with no straw incorporation, the soil solution after 3 wk of flooding showed a negative SI (ranging from –2.4 to –0.8) and after about 6 wk of flooding, positive SIs (ranging from 0–0.9) were obtained. In contrast, positive SI of FeS was computed for both levels of straw treatments (from 0.41–0.96 for 6 g straw kg–1 soil and from 0.97–1.25 for 23 g straw kg–1 soil treatment) within 3 wk of flooding. The changes in Fe(II) concentration (Fig. 5) and decreases in sulfate concentration (Fig. 6) also indicate the formation of FeS. However, these changes do not correspond to the accumulation of sulfide in the solution phase (Fig. 7), which strongly indicates that precipitation of sulfides is playing a major role in regulating sulfide solubility. Positive SIs were also computed for another sulfide mineral—mackinawite (Fe9S8)—whenever positive SIs were computed for amorphous FeS. However, the role of mackinawite in regulating the sulfide solubility is not certain since it was recognized as a diagenetic conversion form from amorphous FeS (Doner and Lynn, 1989). The primary role of amorphous FeS in controlling sulfide concentration has been reported in the literature (Yoshida, 1981; Neue and Bloom, 1989). Although no thermodynamic data are available for MnS formation, Mn(II) may coprecipitate with FeS (Morse et al., 1987). Using WATEQ, at a level of 
0.2 mmol/L Fe(II), precipitation could occur when sulfide concentration reaches about <0.001 mmol/L as a positive SI of FeS was calculated.
Speciation results from WATEQ also indicate possible precipitation as FeCO3 (siderite) and MnCO3 (rhodochrosite) as positive SIs were computed, and these reactions may play certain roles in reducing the concentrations of Mn and Fe in the solution phase. The positive SIs for the carbonates were computed either simultaneously or after positive SIs for FeS were computed, indicating that initial Fe(II) concentration decrease was largely due to formation of FeS.
Analytical results on pore waters (Fig. 4–7) have shown that sulfate concentrations dropped to below detection limit around 13 September for both straw treatments (6 and 23 g straw kg–1 soil). At the same time, both Mn(II) and Fe(II) concentrations approached a relatively low level, and correspondently higher sulfide concentrations were observed for treatments with no straw but with sulfate additions. This in some way supports our hypothesis that precipitation of sulfide minerals controls sulfide solubility. The highest sulfide concentrations observed in the no-straw treatment were probably due to less amount of Fe(III) reduction to Fe(II), which forms less precipitate although this observation was not consistent with the adverse impact of straw treatment on rice that was observed in much early growing stages. Literature has concluded that younger rice plants are much more sensitive to sulfide than they are at later growing stages. One should note that the scales of these figures are different, and some changes in sulfate concentrations could not be visualized. Nonetheless, data have indicated sulfide solubility was affected by redox reactions in paddy soil. However, the effect of kinetics of these processes on sulfide solubility bears further study to understand more comprehensively the interactions of redox reaction on sulfide solubility.
Increases in dissolved methane concentration indicate methanogenesis also occurred early on (Fig. 8), and it is strongly affected by the levels of straw incorporation. There were peaks in methane production that occurred about 6 wk after flooding for treatments with straw. The higher straw addition resulted in a higher peak concentration of dissolved methane, about 0.2 mmol L–1 CH4 with 23 g straw kg–1 soil compared with 0.1 mmol L–1 with 6 g straw kg–1 soil. For the treatment with no straw, dissolved methane concentrations were very low (<0.02 mmol L–1) for up to 11 wk and eventually increased to 0.05 mmol L–1 by the end of the experiment. The peaks observed earlier at about 6 wk after flooding are believed to be related to straw decomposition, and the peaks observed later are related to the excretion of rice root exudate (Sass et al., 1991; Sass and Fisher, Jr., 1997). Methane production, either from CO2 reduction or fermentation, is favored by high organic levels. The results shown in Fig. 2 through 8 indicate overlaps in redox processes.
The dominant terminal electron acceptors by sampling date identified were summarized in Table 2. The redox reactions generally followed the sequence of reduction of oxygen, nitrate, Mn oxyhydroxides, Fe oxyhydroxides, sulfate, and CO2. However, overlapping of electron acceptors, typically among Mn, Fe, and SO2–4, occurred for treatments with no straw addition and among Mn, Fe, SO2–4, and CO2 for 6 g straw kg–1 soil treatments. For 23 g straw kg–1 soil treatments, CO2 became the dominant terminal electron acceptor in about 3 wk after flooding. The overlapping among electron acceptors was also observed in field studies, and this phenomenon had been examined and discussed in depth for this paddy soil (Gao et al., 2002). Compared with the field results, the terminal electron-accepting processes are much more distinct in the greenhouse study, possibly due to the more homogeneous soil conditions and treatments imposed.
Geochemical Redox Classes by Oxidative Capacity
Table 2 also contains the geochemical redox classes (oxic, postoxic, sulfidic, and methanic) of the pore waters based on OXC computation. The calculation and classification based on OXC components were described in Gao et al. (2002). In brief, the oxic status is related to presence of oxygen. Postoxic conditions are related to slightly reduced conditions such as nitrate, Mn, and Fe reduction processes. Sulfidic status is dominated by sulfate reduction, and methanic condition reflects methanogenesis. By using certain criteria in comparing the OXC components, a geochemical classification is realized as a result of integrating all of the oxidized and reduced species in the solution. The OXC required an estimate of bioavailable Mn and Fe oxyhydroxides for which analytical methodology is not available. Since extractable amorphous Mn and Fe oxyhydroxides appeared unreasonably high, the maximum product concentration of Mn and Fe was used as a close estimate of bioavailable Mn and Fe electron acceptors (Gao et al., 2002).
The pore waters for no-straw treatment were oxic to postoxic throughout most of the study, with a final methanic condition identified at the last sampling date. Soil pore waters for the 6 g straw kg–1 soil treatments were identified as sulfidic at 3 wk after flooding and then methanic through the rest of the study. Pore waters in the 23 g straw kg–1 soil treatments were identified methanic as early as 18 July, about 3 wk after flooding. Since complete chemical analysis data were not available before 18 July, we do not know if methanic conditions occurred earlier for the treatments receiving straw. Clearly, more reducing conditions progressed at a much faster rate in the higher straw treatments compared with the lower and no-straw treatments.
Rice Plant Response to Straw and Sulfate Treatments
Rice plants responded strongly to the straw treatments in the early growing stages. Plant height at about 4 wk and number of tillers at about 6 wk after planting were both reduced greatly by straw incorporation (Tables 3 and 4). The average height of plants in the 23 g straw kg–1 soil treatment was much shorter than the 0 and 6 g straw kg–1 soil treatments. The average number of tillers varied among both the straw and sulfate treatments. Grain yield decreased significantly as quantity of straw incorporated increased (Tables 3 and 4). Sulfide formation, as evidenced by the rotten-egg smell during the course of the experiment, was noted in the straw incorporation treatments. It was further observed that maturation of rice plants in the highest straw (23 g straw kg–1 soil) treatment was delayed with many more sterile panicles formed compared with other treatments. These symptoms are consistent with the sulfide toxicity symptoms reported in other studies (Kuo and Mikkelsen, 1981; Allam and Hollis, 1972).
F tests show that grain yield variation associated with straw treatment effects were significant (p < 0.0001), but the variations in grain yield due to sulfate addition and the interaction between straw and sulfate addition were not significant (Table 4). Orthogonal contrasts among treatments show that the differences in grain yield between no-straw treatment and treatments with straw (0 vs. 6 and 23 g straw kg–1 soil) as well as between the two straw levels (6 vs. 23 g straw kg–1 soil) were both significant (p < 0.0001). Similar conclusions were drawn for straw treatment effect on plant height at 4 wk and the number of tillers at 6 wk after planting (p = 0.005 to <0.0001). When analyzing the effect of sulfate addition, however, only the difference between treatments with and without sulfate additions (0 vs. 160 and 800 mg SO4 kg–1 soil) was significant (p = 0.0028) on tillers. The differences on tillers between the two levels of sulfate addition as well as on both grain yield and plant height were not significant.
Based on the observations in this study, sulfide toxicity to rice plants could be one of the causes in rice biomass reductions from straw incorporation as sulfide toxicity symptoms were observed. However, direct evidence of high soluble sulfide concentration associated with these symptoms was not found. Potential toxicity of Fe(II) and Mn(II) was excluded in our study because the concentration levels observed were in the normal range and rice plants did not show characteristic metal toxicity symptoms (Tadano and Yoshida, 1978). In addition, organic acids (e.g., formic, acetic, propionic, and butyric acids) are reportedly toxic to rice at concentrations of 0.01 to 0.001 M (Tadano and Yoshida, 1978). The production of organic acids is enhanced in the soil with readily decomposable organic matter such as rice straw. We found greater than 200 mg L–1 dissolved organic C within the first 3 wk of flooding in the 23 g straw kg–1 soil treatment but are not certain if toxic levels of organic acids were produced at this level of dissolved organic C.
This greenhouse study was conducted at slightly higher salinity than normal field conditions because initial EC in the soil was 2.5 dS m–1 and addition of sulfate or other equivalent salts for the treatments was imposed. However, sulfate addition did not impact plant biomass reduction except that a significant reduction in the number of tillers in early growing stage was observed. The sole factor in reducing plant biomass appeared to be straw incorporation, and the impact on rice plants appeared much more sensitive in early growing stages. The treatment of 6 g straw kg–1 soil incorporated (equivalent to 6 tons ac–1) in this greenhouse study was higher than the average return rate (4 g straw kg–1 soil). Under field conditions, only uneven incorporation of straw may result higher incorporation rate in localized areas. In addition, sulfide toxicity appears to occur at the drainage outlet areas, i.e., associated with stagnant water and higher salinity (containing sulfate ions). Thus, the problem may be exasperated by increasing soil and water salinity (sulfate in particular) due to reduction in diversion of supply water from the Sacramento River to protect salmon fry from being sucked into the pumps and the requirement to hold flood waters for specified time period after chemical application in paddy fields to reduce chemical residues in drain waters (Tanji and Watanabe, 2000). Both of these constraints have resulted in extensive drain water reuse and rise in water and soil salinity (Scardaci et al., 1996).
|
CONCLUSION
|
|---|
A greenhouse study was conducted to more clearly define the environmental conditions leading to the random occurrence of sulfide toxicity in paddy rice grown in the Sacramento Valley. Sulfide toxicity symptoms were observed when excessive amounts of straw (one-time incorporation of 6 and 23 g straw kg–1 soil) were incorporated into the soils and in the presence of moderate concentrations of sulfate. Sulfide is formed from the reduction of sulfate under sulfidic redox conditions, but it tends to precipitate out as ferrous and possibly manganous sulfide, and as a result, low range of sulfide concentrations were detected in soil solutions. Plant response, chemical analyses, and chemical speciation modeling all support that sulfide toxicity can occur from straw incorporation under specific environmental conditions. Under current practices of straw incorporation, about 4 g straw kg–1 soil, sulfide toxicity is not expected to be observed on a large scale.
|
ACKNOWLEDGMENTS
|
|---|
Funding for this research was provided by Kearney Foundation of Soil Science, Division of Agriculture and Natural Resources, University of California.
|
REFERENCES
|
|---|
- Allam, A.I., and J.P. Hollis. 1972. Sulfide inhibition of oxidases in rice roots. Phytopathology 62:634–639.[ISI]
- Ball, J.W., D.K. Nordstrom, and D.W. Zachmann. 1987. WATEQ4F: A personal computer FORTRAN translation of the geochemical model WATEQ2 with revised data base. Open File 87–50. U.S. Geological Survey, Denver, CO.
- Doner, H.E., and W.C. Lynn. 1989. Halide, sulfate, and sulfide. p. 279–330. In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. SSSA Book Ser. 1. 2nd ed. SSSA, Madison, WI.
- Gao, S., K.K. Tanji, S.C. Scardaci, and A.T. Chow. 2002. Comparison of redox indicators in a paddy soil during rice-growing season. Soil Sci. Soc. Am. J. 66(3):805–817.
- Kuo, S., and D.S. Mikkelsen. 1981. The effects of straw and sulfate amendments and temperature on sulfide production in two flooded soils. Soil Sci. 132:353–357.
- Morse, J.W., F.J. Millero, J.C. Cornwell, and D. Rickard. 1987. The chemistry of the hydrogen sulfide and iron sulfide systems in natural waters. Earth–Sci. Rev. 24:1–41.
- Neue, H.U., and P.R. Bloom. 1989. Nutrient kinetics and availability in flooded rice soils. p. 173–190. In Progress in irrigated rice research. IRRI, Manila, Philippines.
- Patrick, W.H. 1981. The role of inorganic redox systems in controlling reduction in paddy soils. p. 107–230. In Proc. Symp. on Paddy Soil, Nanjing, China. 19–24 Oct. 1980. Sci. Press, Beijing, and Springer-Verlag, New York.
- Ponnamperuma, F.N. 1981. Some aspects of the physical chemistry of paddy soil. p. 59–94. In Proc. Symp. on Paddy Soil, Nanjing, China. 19–24 Oct. 1980. Sci. Press, Beijing, and Springer-Verlag, New York.
- SAS Institute. 1991. SAS system for elementary statistical analyses. SAS Inst., Cary, NC.
- Sass, R.L., and F.M. Fisher, Jr. 1997. Methane emission from rice paddies: A process study summary. Nutr. Cycling Agroecosyst. 49:119–127.
- Sass, R.L., F.M. Fisher, P.A. Harcombe, and F.T. Turner. 1991. Mitigation of methane production and emission in a Texas rice field. Global Biogeochem. Cycles 4:47–68.
- Scardaci, S.C., A.U. Eke, J.E. Hill, M.C. Shannon, and J.D. Rhoades. 1996. Water and soil salinity studies on California rice. Rice Publ. 2. Cooperative Extension, Colusa, CA. Available online at http://agronomy.ucdavis.edu/uccerice/WATER/salinity.htm) (Verified 12 Sept. 2003.)
- Tadano, T., and S. Yoshida. 1978. Chemical changes in submerged soils and the effect on rice growth. p. 399–420. In Soil and rice. IRRI, Los Baños, Philippines.
- Tanji, K., and T. Watanabe. 2000. Environmental and ecological constraints on rice irrigation in California, USA. p. 91–96. In Proc. Asian Regional Workshop on Sustainable Dev. of Irrig. and Drain. for Rice Paddy Field, Tokyo. 24–28 July 2000. Jpn. Natl. Committee on Irrig. and Drain., Tokyo.
- Yoshida, S. 1981. Fundamentals of rice crop science. IRRI, Los Baños, Philippines.
TOP
ABSTRACT
INTRODUCTION
