| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Hydrology Program, Dep. of Land, Air and Water Resources, Univ. of California, Davis, CA 95616
b Dep. of Agronomy and Range Science, Univ. of California, Davis, CA 95616
* Corresponding author (sugao{at}ucdavis.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
Abbreviations: DO, dissolved O2 • EH, redox potential • ERL, electron reference level • GC, gas chromatography • OXC, oxidative capacity • PVC, polyvinyl Cl • RSD, relative standard deviation • SB-WF, straw burned with winter-flooding • SR-NWF, straw rolled with no winter-flooding • SR-WF, straw rolled with winter flooding • TEAPs, terminal electron-accepting processes • UC, University of California • UHP-N2, ultra-high purity N
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
The objective of this study was to evaluate three redox indicators: conventional EH measurement, dominant TEAPs, and OXC in paddy soil. The paper first describes the three redox indicators, presents details on field sampling technique since this is our first attempt of applying TEAPs and OXC to paddy soil, and then evaluates the three approaches to characterizing redox status.
|
Description of Redox Parameters |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
, Mn(IV, III)/Mn(II), Fe(III)/Fe(II), 
, and CO2/CH4. The reduced conditions in submerged paddy soils may be readily measured by measuring EH of the pore water, but EH is a difficult soil parameter to interpret. Bartlett (1999) described thoroughly the redox behavior in soils, providing an important base to our understanding of equilibrium and dynamic redox conditions. He pointed out that a Pt electrode may not reflect changes in some species involved in redox reactions, such as partial pressure of O2 and neither Mn or Fe oxides nor nitrate had the expected quantitative effect on the Pt electrode measurement. Methane, bicarbonate, N2 gas, nitrate, and sulfate are not electroactive, i.e., they do not readily take up or give off electrons at the surface of the Pt electrode used to measure EH (Berner, 1981). Since it is a measurement of potential, the Pt electrode also responds to changes in pH and other potentials. Thus, measured EH usually reflects a nonequilibrium mixed potential and can be only qualitatively interpreted (Bohn, 1971). Indeed, a wide range of EH for the same redox couples or several redox reactions occurring within the same range of EH were reported throughout literature (Lovely and Goodwin, 1988). Under experimentally controlled EH conditions, sequential reduction of nitrate and Mn(IV) or sequential oxidation of Mn(II) and NH+4 has been determined (Patrick and Jugsujinda, 1992). However, redox reactions in most natural systems, especially soils, are seldom at equilibrium. Therefore, nonequilibrium and capacity-type redox parameters are desired to better assess reduced soil conditions.
Terminal Electron-Accepting Processes
For the nonequilibrium approach, defining the dominant TEAPs has been applied to groundwater systems to predict the predominant redox reactions under anoxic conditions (Chapelle et al., 1995). As shown in Fig. 1
, the identification of dominant TEAPs along the flow path in a groundwater system (Chapelle et al., 1995) considers simultaneously the consumption of electron acceptors (DO, NO-3-N, Fe(III), SO2-4-S, and CO2); intermediate product (dissolved hydrogen gas, H2) concentration, and the concentrations of final products (Fe(II), H2S, and CH4). It should be noted that Mn(IV, III) reduction was not included in Fig. 1 which will be considered above Fe(III) reduction in this research. A high degree of confidence can be achieved if a combination of all three indicators (electron acceptor consumption, intermediate product concentration, and final product accumulation) yields a positive identification of the predominant TEAPs. When only two out of three indictors confirm a predominant TEAPs, the diagnosis is less positive.
|
We attempted to apply the same method described by Chapelle et al. (1995) to determine the dominant TEAPs in the pore waters of submerged rice paddies, an environment far different from that of groundwater systems. Not considered was the flow path in the shallow rootzone of the paddy rice but changes with respect to time was chosen instead. A diagnosis may not be completely available in the case of Mn and Fe oxides because they are only sparingly water soluble.
Oxidative Capacity
The concept of OXC was developed by Scott and Morgan (1990), leading to geochemical classes of redox conditions (Berner, 1981). The OXC is a capacity-type parameter that utilizes a comprehensive chemical analysis of the water (oxidized and reduced species) into a single descriptive parameter (Maruyama and Tanji, 1997). Table 1 (Scott and Morgan, 1990) contains the major redox half reactions operating in rice paddies including the number of electrons transferred. For the last half reaction, Scott and Morgan used production of dissolved organic matter which we replaced with the CO2–H2 reduction pathway for methane (Snoeyink and Jenkins, 1980).
|
 | [1] |
Scott and Morgan (1990) further defines OXC as
 | [2] |
It should be kept in mind that Eq. [2] includes only the most common redox elements in soils and many other minor electron acceptors are ignored.
The redox status of the pore water in the paddy soil is then transformed into geochemical redox classes analogous to that by Scott and Morgan (1990):
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
The soil at the field site is Willows clay (fine, smectitic, thermic Sodic Endoaquert). This soil type is widely distributed in Colusa and Glenn Counties, California and is considered the main rice soil in the area.
Standard grower practices were used in the field, i.e., continuous flooding except for short-term drainage for broadleaf weed control. Flooding was initiated on 25 May 1998 and completed on 26 May. The plots were seeded into 10 cm of water on 27 May to variety M-202, a medium grain rice. The plots were drained on 7 July for spraying a mix of herbicide and were reflooded 3 d later to 18- to 20-cm water depths. Such a brief drain period only allowed surface water to drain and increased the pesticide contact to the surface soil. Sampling was not made during the drainage dates. The rice paddy was eventually drained on 11 September for harvest in early October.
Field Instrumentation
Piezometers (sampling wells) were installed in the selected plots at one-third of the distance from the drain prior to flooding on 24 May 1998. Figure 2
shows a sketch of the piezometer installation and expected direction of water flow while sampling water from the piezometer. The piezometers were 5-cm diam. polyvinyl Cl (PVC) pipes with a plastic screen (
0.5-mm sieve size) at the bottom. Studies with PVC-casted wells have shown that no H2 is produced from this material (Chapelle et al., 1997). The piezometers were installed 5 cm below the soil surface. A hole 10 cm deep and about a 30-cm diam. was made before each piezometer was installed. By placing the piezometer bottom to the 5-cm depth with the support of three bamboo sticks glued on the piezometer bottom, the empty space below the piezometers and the surroundings was back-filled with quartz sand. The sand served as a filter to obtain clear solution samples from the soil pore water. The PVC pipe was equipped with a threaded cap with an acrylic tube inserted from the top, long enough to reach the piezometer bottom. Finally, a 30-cm PVC ring was used to surround the piezometer. A layer of 5 cm deep of bentonite was placed inside the 30-cm PVC ring surrounding the piezometers to prevent surface water building up around the piezometer and from leaking through while sampling the soil pore water from the piezometer. A cap was placed on the PVC ring.
|
Sampling
A preliminary test showed that a vacuum pumping method was adequate to sample large volumes of the pore water from the PVC piezometers. A hand-operated vacuum pump (Nalgene, Nalge, Rochester, NY) was used to extract water out of the piezometers through Tygon tubing. Using the hand pump, a continuous water stream flow (150–200 mL min-1) could be controlled easily without introducing any air bubbles into the line. Initially, a Markson ORP Pt/Ag/AgCl combination electrode (Markson Science Inc., Phoenix, AZ) was lowered down to the bottom of the piezometer to monitor EH while pumping and later a flow-through chamber was developed for monitoring both DO and EH on line while sampling.
The time to collect water samples from the piezometer was based on preliminary tests on changes in water parameters as shown in Fig. 3 . EH readings became relatively stable after a 2-L volume of water pumping. Changes in DO, NO-3-N, Mn(II), Fe(II), and DOC (dissolved organic C) were all very small after 1 or 1.5 L water pumping. Dissolved H2 gas and methane in soil pore water were tested for two water samples at 1.5 and 4.5 L volume of water pumping and the variations between the two readings were small. The only change noticeable during water sampling was SO2-4-S concentration after a 2 L water pumping and the decrease was exponential. Water samples were not collected for chemical analysis until the EH readings became relatively stable and after pumping volume reached 2 L for the piezometers.
|
Several water samples were collected for analyses. Unstable parameters such as DO, pH, EH, and S2- were measured on-site using fresh water samples as it is critical to minimize sample exposure to the air before these analyses. A 125-mL Nalgene polyethylene bottle (Nalge Co., Rochester, NY) was filled up with water sample and stored in a cooler with ice for later lab analysis for NO-3, SO2-4, and EC. Another 125-mL Nalgene bottle of sample was filled up and preserved in 1% (vol./vol.) HCl for analysis of Fe(III/II) and Mn(II). It was found out later that these samples should be filtered first (0.1 µm or smaller) prior to acidification for Fe(III/II) analysis to prevent dissolution of colloidal Fe oxyhydroxides. Otherwise, an overestimation of Fe(III) could result.
For dissolved H2, and CH4 analysis, evacuated 400-mL PVC bags (Medsep Corp., Covina, CA) were filled up with water sample. The PVC bags had two ports and one was with a septum (Alltech Inc., Deerfield, IL). Samples were introduced into the bag through the other port. These bags were weighed before and after filling to obtain the mass of sample.
After collecting the water samples into PVC bags, a known amount (10 mL) of ultra-high purity N (UHP-N2) was injected into the bags. After equilibrating the N2 gas phase with the liquid phase (minimum 7 min) (Chapelle et al., 1995), 5 mL of the N2 gas phase was withdrawn using a syringe and injected into 10-mL vacutainers previously flushed with UHP-N2. The vacutainers were later used for H2 and CH4 analysis by gas chromatography (GC). Standards of H2 and CH4 were prepared in the same way as the samples using the vacutainers to minimize calculation error on dilution factors. A blank of UHP-N2 was made for background correction. The calculations for dissolved gas concentrations (H2 and CH4) were based on concentrations in the gas phase, dimensionless distribution coefficients for aqueous-gas equilibrium (Stumm and Morgan, 1981), and the volumes of gas and liquid phases. This gas sampling procedure was developed based on the stability test conducted as described below.
The stability of H2 gas in the PVC bags was tested within 24 h after filling the bags with a known concentration of H2 gas. The stability of H2 and CH4 in the vacutainers was also tested. The vacutainers were flushed with UHP-N2 in a glove-bag five times and then 5 mL of standards of H2 or CH4 was injected into the vacutainers. The concentrations of these gases were 11.0 µL L-1 for H2 and 511 µL L-1 for CH4. One group of the vacutainers was kept in a dark plastic bag at room temperature (22–23°C) and the other group was kept in the refrigerator (5°C). Five replicates of each treatment were set up and the concentrations of H2 and CH4 in the vacutainers were analyzed by GC within 48 h.
The results of the recovery test of gaseous stability in PVC bags and vacutainers are briefly summarized here. For the PVC bags, recovery of H2 gas was 100% within 6 h and 90% after 24 h. For the vacutainers, recovery of H2 gas concentration in the vacutainers after preserving for 20 h were >98% at both room temperature and refrigerated conditions with a standard deviation 
5%. After preserving and holding for 72 h, the recovery was >98% with a relative standard deviation (RSD) of 
6.5%. For CH4, the recovery after 48 h ranged from 95 to 97% with a RSD of 2 to 5%. There were no significant differences between recoveries from vacutainers kept at room temperature and refrigerated condition. Thus, recovery of 95% of H2 and CH4 can be achieved within 48 h without exposure to light either at room temperature (23°C) or cooler conditions. By preparing the samples in the field and preserving the gas phase in UHP-N2-flushed vacutainers, the work load can be greatly reduced in the field.
Analytical Methods
The pH was measured with a Piccolo stick pH meter (Fisher Scientific, Santa Clara, CA). The EH was measured using a Sensorex ORP Pt/Ag/AgCl combination electrode. The ORP electrode was calibrated using Zobell's solution (Nordstrom, 1977). To standardize the EH reading to the standard H2 electrode, 198 mV was added to the observed instrument reading to obtain the standard EH values (Nordstrom, 1977). Dissolved O2 values were based on the measurement with YSI model 54A oxygen meter DO probe (Yellow Springs Instrument Co., Yellow Springs, OH). Sulfide was determined using a CHEMtrics analysis kit (CHEMtrics Inc., Calverton, VA) based on a colorimetric method (USEPA, 1983) and the detection limit was 0.1 mg S2--S L-1. Specific conductance (EC), NO-3-N, and SO2-4-S were measured in the lab within 24 h after sampling using the preserved samples. The analysis for NO-3-N was conducted using the Brucine-method (Barker, 1974). Iron (II) and Fe (III) were measured using the ferrozine method (Stookey, 1970) and Mn (II) concentrations were determined using atomic absorption spectroscopy (AAS). To estimate the amount of Fe and Mn potentially serving as electron acceptors of the soils, amorphous or "active" Fe and Mn oxides were extracted by ammonium oxalate buffer (pH 3.0) (Loeppert and Inskeep, 1996).
The concentration of dissolved H2 was determined with a RGA3-gas analyzer with a Reduction Gas Detector (Trace Analytical Lab, Menlo Park, CA). Parameters used were carrier N2 gas flow of 20 mL min-1 measured from the outlet of Hg scrubber column, detector temperature of 265°C, and column temperature of 105°C. The detection limit of this instrument was 10 µl L-1 H2 and detection limit for the solution was 0.2 nmol L-1 based on the sample preparation procedures described. Methane was determined with a SRI 1860 Gas Chromatograph (SRI Instruments, Torrance, CA) equipped with flame ionization detector (FID).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
|
|
The concentration of DO and NO-3-N for all plots decreased rapidly corresponding to the initial EH value drop. However, under anoxic conditions, even though EH varied little, the concentrations of reduced products such as Mn(II), Fe(II), and CH4 continuously increased while SO2-4-S continuously decreased. The results confirm the difficulty in using measured EH readings to indicate redox reactions occurring in the soil especially under anoxic conditions.
Deducing Dominant Terminal Electron-Accepting Processes
Changes in Electron Acceptors and Products
Dissolved O2 and NO-3-N.
Dissolved O2 dropped from >0.2 to <0.02 mmol L-1 after 5 d of flooding and then remained below 0.02 mmol L-1. The changes in the concentrations of NO-3-N followed almost the same pattern as that of DO. The concentration of NO-3-N dropped from 1.8 to <0.1 mmol L-1 within the first 5 d of flooding. Clearly, O2 and NO3 were the dominant terminal electron acceptors within the first 5 d or so after flooding for all the plots.
Manganese and Fe.
Since Mn and Fe oxides are sparingly soluble and not detectable, increases in Mn(II) and Fe(II) concentrations indicates Fe and Mn were serving as electron acceptors. Upon flooding, there was a steady increase of Mn(II) for all the plots indicating that Mn oxides (Mn IV, III) in the solid phase were serving as an electron acceptor very early in the growing season. It should be mentioned that the soil was disked several times for seed bed preparation and thus it is expected that Mn was in an oxidized form before flooding. The concentrations of Mn(II) eventually leveled off or decreased, possibly because of depletion of bioavailable Mn oxides and precipitation as MnS. In contrast, the concentrations of Fe(II) in the soil pore water in all three plots did not increase abruptly until late June, about 4 wk after flooding, with the exception of SR-WF treatment which showed an abrupt increase after 5 d of flooding, then leveled off and increased again, indicating Fe(III) in the plot was serving as an electron acceptor at earlier times.
Sulfate and Sulfide.
The initial concentrations of SO2-4-S were highly variable among the treatments, partly because of the differences in water salinity in each plot (Fig. 5). Sulfate concentrations decreased rapidly from >6 to 0.2 mmol L-1 in late July, 8 wk after flooding. The possibility of decrease in SO2-4-S because of precipitation as gypsum was rejected because the soil solution Ca concentration (not shown) was below 0.25 mmol L-1. Using speciation model MINTEQA2 (Allison et al., 1991), soil pore waters during most of the growing season were undersaturated with respect to gypsum (Ksp = 10-4.6, Allison et al., 1991). The SR-NWF and SR-WF treatments showed a steady decrease in SO2-4-S concentration while SB-WF plot showed some early fluctuations in concentration and then gradually decreased. Sulfate seemed to serve as an important electron acceptor throughout the growing season for the SR-NWF and SR-WF Plots after flooding and for SB-WF Plot it became important after a few weeks of flooding. Sulfide was also determined in the pore waters but was all below the detection limit of 0.1 mg L-1. Based on Fe(II) concentration in pore waters, the theoretical S2- concentration in the pore waters for solutions saturated with FeS (Ksp = 10-3.915) would be <0.1 mg L-1 according to MINTEQA2 (Allison et al., 1991). The analytical method used in this study was not able to detect low concentrations of S2-. Thus, direct evidence of S2- accumulation in soil solution was not observed.
Methane.
In the SR-NWF Plot, dissolved CH4 concentration increased dramatically after 17 June and reached peak concentration on 24 July and leveled off thereafter. For the other two treatments, the highest concentrations of CH4 were determined on the last sampling date, 3.5 mo after flooding. The SR-WF plot was flooded during the winter and decay of rice straw was promoted more than no winter-flooding plot and that might contribute to the smaller methane production. Methane is a product of anaerobic decomposition of soil organic matter and is evidence of strongly reducing conditions.
Intermediate Product-Dissolved Hydrogen Gas Concentration
The concentrations of dissolved H2 varied between 0.6 to 3.2 nmol L-1 (Table 2). The range of dissolved H2 found in this study fell into sulfate reduction process based on the criterion provided by Chapelle et al. (1995) although other TEAPs were occurring during the sampling period.
|
|
|
Using the maximum concentrations of reduced Fe(II) and Mn(II) may underestimate the amount of Fe and Mn available as electron acceptors because of possible precipitation such as FeS and MnCO3. The alkalinity of soil pore water was monitored in the previous year and averaged 11, 14, and 12 mmol L-1 as HCO-3 for SB-NW, SR-NWF, and SR-WF, respectively. Based on these levels, the pore waters were undersaturated with respect to FeCO3 but oversaturated with respect to MnCO3. Precipitation such as FeS and MnCO3 would remove Fe(II) and Mn(II) from the solution phase. It was observed later in a greenhouse study (Tanji et al., 2001) that the maximum concentration of Mn(II) and Fe(II) in the soil pore water was affected by the amount of incorporated straw. The higher amount of straw incorporation resulted in the earlier appearance of the maximum concentration of Fe during the rice-growing season which was also accompanied by higher concentrations of soluble Fe(II). The highest Fe(II) concentration observed was 1.4 mmol L-1 when 2.3% (w/w) straw was incorporated and 0.2 mmol L-1 concentration of Fe(II) when no straw was incorporated. A good estimate on the amount of Fe and Mn serving as electron acceptors should fall between the maximum concentration of Fe(II) and Mn(II) observed up to 1.4 mmol L-1 for this field. Accurate determination on bioavailable Fe and Mn oxyhydroxides serving electron acceptors using laboratory technique bears further testing.
Therefore in this study, estimates of bioavailable Mn oxides were taken as 0.18, 0.16, and 0.20 mmol L-1 and bioavailable Fe oxides as 0.12, 0.14, and 0.19 mmol L-1 for the SB-NW, SR-NWF and SR-WF plots, respectively. These values represent the maximum concentration of Mn(II) and Fe(II) resulted from reduction of Mn and Fe oxides during the rice growing season. In all the three plots, oxic and postoxic status are found throughout most of the rice-growing season. Oxic conditions were determined during early growing season. Then, postoxic conditions were observed through most of the growing season. Sulfidic conditions were identified on 9 September for both SR-NWF and SR-WF plots in which straw was incorporated. Methanic conditions were not identified for any of the plots.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
, and as examples) at pH 7.0 using speciation model WATEQ (Ball et al., 1987). The results are plotted in Fig. 6
. It is important to notice that the computation was based on assumptions that the solution was dilute enough so that most redox species are in free ion forms and no precipitation such as FeS was involved as the ratios of redox couples change. Nitrate reduction occurs at high redox potential within a narrow range (300–350 mV) considering NO3/NH4 ratio change from 106 to 10-6. The same applies for sulfate reduction except at a much lower redox potential range (-200 to -250 mV). For the Fe3+/Fe2+ redox couple, however, reduction or oxidation could occur in a much wider range of EH depending on the nature of the reactions (Fig. 6). The slope of EH change in Fe3+/Fe2+ redox couple was about eight times steeper than that of 
and because of the different number of electrons transfer involved in the reactions. The same would occur for Mn which involves approximately one electron transfer as in Fe redox reactions, covering a wider range of EH. Thus, the possibility of electron overlap exists but the degree of the overlapping depends on specific conditions such as kinetics and precipitation.
|
Moreover, the concentration of reduced products may be affected by mineral solubility such as MnS, FeS, and MnCO3, etc, which removes soluble Mn, Fe, and H2S (or HS-) from the solution phase. Precipitation tends to increase Fe3+/Fe2+ ratios and drives the reduction reaction forward. Theoretically, precipitation occurs once the ionic activity product of the mineral is exceeded. Postma and Jakobsen (1996) provided an explanation for the simultaneous occurrence of Fe(III) and SO4 reduction using the partial equilibrium approach. They stated that depending on the stability of Fe oxides, simultaneous reduction of Fe(III) and SO4 is thermodynamically possible under a wide range of sedimentary conditions and SO4 reduction may even occur before Fe(III) reduction. The presence of a wide range of Fe oxide stability is likely to cause considerable overlap between zones of Fe(III) reduction and SO4 reduction.
Dissolved Hydrogen Gas Concentration as Indicator of Terminal Electron-Accepting Processes in Paddies
According to Chapelle et al. (1995), as reducing conditions develop and less electrochemically positively electron acceptors become dominant, dissolved H2 concentrations should increase in deep-seated ground water systems. In this shallow rice paddy, however, there was no evident trend of increasing dissolved H2 concentrations as more reducing conditions developed.
The observed dissolved H2 concentrations in paddy soils are difficult to evaluate. The dissolved H2 concentrations are mediated by the physiologic characteristics of microorganisms in the system (Lovley and Goodwin, 1988). Overlapping among the TEAPs indicate a mixed community of microorganisms in the paddies that involved Mn, Fe, and sulfate reductions and methane production. The high concentration of dissolved H2 observed while Mn and Fe reductions prevailed could be because of the overlap among electron acceptors, especially the early occurrence of sulfate reduction. On the other hand, paddy soil is a highly dynamic and complex system. During flooding, the soil environment is encountering rapid consumption of O2 as organic matter is oxidized following by progressive development of reduction processes in the paddy with significant overlap indicating constant changes in microbial community.
The rootzone of paddy soil is very shallow, 15 cm in depth with a much greater soil-water-air interface as compared with deeper groundwater systems in which the TEAPs method has been previously applied. Thus, any possible diffusion in the submerged paddy soil may impact the dissolved H2 gas concentrations observed in the pore waters. Further, the TEAPs diagnosis is based on steady-state H2 and this may have been difficult to achieve in paddy soils compared with deep-seated ground-water systems. Therefore, the current results indicate that dissolved H2 concentration may not be suitable for indicating dominant TEAPs for paddy soils where overlap among electron acceptors occurs.
Comparison between Redox Status Indicators
Redox Potential and Dominant Terminal Electron-Accepting Processes
Figure 7
plots measured concentrations of redox species against observed EH and pH values. Since the pH of the pore waters was near neutral, Fig. 7 reflects the dominant TEAPs occurring as redox potential varies from this study. Data from all three experimental plots showed the same trend and only one plot is shown here as an example. Dissolved O2 and NO-3 served as electron acceptors at EH 350 to 400 mV since their concentrations dropped to about zero at 350 mV. Manganese and Fe served as electron acceptors starting around 350 mV for Mn and 250 mV for Fe to 100 mV. Sulfate reduction occurred from EH as high as 350 mV to 100 mV. Methane production, however, was not obvious until EH was close to 100 mV. Manganese, Fe, and SO2-4 reduction processes were occurring over a much wider range compared with O2 and NO-3 reduction and methane production for this soil. These EH values or ranges may vary somewhat from one plot to another but the same phenomenon and trend were observed.
|
, dissolved S concentration was all below detection limit (0.1 mg L-1) using the current analytical method. A more sensitive analytical method is needed for detecting H2S in the solution. Figure 8 plots measured redox species and observed EH corrected to pH 7.0 for all three plots. A higher correlation was obtained for Mn(II), Fe(II), and CH4 with r2 values of 0.76, 0.73, and 0.76, respectively compared with those for DO, NO3, and SO4 (r2 = 0.53, 0.37 and 0.16, respectively). These results indicate that measured redox potential could be explained by the concentration of some redox species. The results basically agree with the literature that DO, NO-3, and SO2-4 are electrode insensitive. Generally speaking, measured redox potential cannot be used to predict the ratio of redox species or couples unless the reactions occur at the metal electrode surface quickly and reversibly (Kölling, 2000). As most of redox reactions in nature especially in rice paddies are not reversible, it is expected that poor correlation exits between measured EH and some redox species.
|
|
Straw Treatment Effect on Redox Status
The incorporation of rice straw results in more reducing conditions in rice field than straw-burned (Fig. 4, Table 3 and 4) as indicated by the lower EH values, the earlier occurrence of more reducing redox couple reactions, and more reducing redox status classified by OXC. The degree to which reducing conditions develop is dependent upon the presence of soil organic matter. The plot with straw burned with winter-flooding produced the least reducing conditions, most likely because of the lower amount of organic C source (straw incorporated). The other two plots that received straw all resulted in lower EH values, higher Mn(II), Fe(II), and CH4 concentrations, and more sulfidic conditions. Comparison among the treatment on soil redox status was limited because of the lack of replicates in this study. However, our further study on straw treatment effects on paddy soil redox chemistry have validated the above results which will be reported in a near future publication.
|
SUMMARY |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 12, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONDescription of Redox ParametersMATERIALS AND METHODSRESULTSDISCUSSIONSUMMARYREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Yang, L. Yang, and L. Jianhua Organic Phosphorus Fractions in Organically Amended Paddy Soils in Continuously and Intermittently Flooded Conditions J. Environ. Qual., May 31, 2006; 35(4): 1142 - 1150. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
