Journal list menu
Visual Assessment of Sulfate Reduction to Identify Hydric Soils
Open Access article
This is an open access article distributed under the CC BY‐NC‐ND license (http://creativecommons.org/licenses/by‐nc‐nd/4.0/).
Abstract
-
Core Ideas
- Iron monosulfides determine the presence of highly reduced conditions.
- Visual sulfate reduction on IRIS surfaces can be used to identify hydric soils.
- Greater S content in saturated soils results in greater monosulfide formation.
Some hydric soils present a unique set of problems associated with their identification, including morphological features not reflective of current hydrology and/or masked redoximorphic features. A simple, reliable tool to identify reduced soil environments is IRIS (Indicator of Reduction in Soil) tubes and panels. In saturated, anaerobic soils with substantial sulfur (S) contents, a commonly observed phenomenon is the reduction of S to form black iron monosulfides (FeS). The objective of this experiment was to document S reduction on polyvinylchloride surfaces coated with synthesized iron (Fe) oxides to simplify the identification of hydric soils wetlands containing moderate to high S concentrations. The IRIS panels were installed in soil mesocosms containing varying concentrations of sulfate and saturated for 11 wk. Oxidation–reduction potential and pH were measured weekly throughout the study. At the termination of the experiment, the IRIS panels were removed and analyzed quantitatively for percentage of oxidized iron (Fe3+) remaining, reduced iron (Fe2+) removed, and reduced S (FeS) precipitated onto the panel. When IRIS panels displayed 2% FeS precipitate, 77% of the panels met the National Technical Committee for Hydric Soil criteria using reduced Fe removal from IRIS panels, and 100% of the panels met the criteria established using measured oxidation–reduction potential. The strong visual observation of FeS is a simple, quick determination of highly reduced conditions. Wetlands are a valuable natural resource that can be challenging to delineate, and the incorporation of a visual S reduction criteria on IRIS surfaces is beneficial for the timely and accurate identification of hydric soils.
Abbreviations
-
- B
-
- blue
-
- Eh
-
- oxidation–reduction potential
-
- G
-
- green
-
- IRIS
-
- Indicator of Reduction in Soil
-
- NTCHS
-
- National Technical Committee for Hydric Soils
-
- R
-
- red
-
- redox
-
- oxidation–reduction
Wetlands are recognized as an important source of biodiversity, recreation, and ecosystem services. The protection and delineation of wetlands is not only ecologically important, but it is also mandated by federal law under Section 404 of the Clean Water Act (21). Wetland delineations encompass three factors: hydrophytic vegetation, hydric soils, and wetland hydrology (20). Seasonal vegetation and fluctuating water tables can alter the presence of vegetative and hydrologic wetland indicators throughout the year. Alternatively, the biogeochemical conditions that result in hydric soil morphologies persist in the soil throughout seasonal variation and environmental modifications, making hydric soils particularly useful in wetland delineation (22).
A hydric soil is defined as a “soil that formed under conditions of saturation, flooding, or ponding long enough during the growing season to develop anaerobic conditions in the upper part” (8). Anaerobic soil environments develop several distinctive properties compared with aerobic environments, including accumulation of organic matter and the chemical reduction and translocation of reducible elements such as Fe and S (22).
Chemical reduction reactions occur as a result of microbial respiration. Microorganisms oxidatively metabolize carbon (C), which requires a reciprocating reduction reaction with a terminal electron acceptor (TEA) (23). Oxygen is the most thermodynamically favorable TEA. As anaerobic conditions develop in saturated soils, oxygen (O2) is rapidly depleted, forcing microbes to couple organic matter oxidation with less thermodynamically favorable TEAs (6). Alternate TEAs, commonly found in soils, in decreasing thermodynamic favor are nitrate (NO3−), manganese (Mn4+), Fe3+, sulfate (SO42−), and carbon dioxide (CO2) (23). Thus, reducing conditions are characterized by the use of TEAs other than O2.
(1)
(2)Oxidized Fe3+ is insoluble and colored, whereas reduced Fe2+ is soluble and colorless. In reducing environments when Fe3+ is reduced to Fe2+, the orange‐colored Fe3+ paint used in the construction of the IRIS tubes is reduced to soluble Fe2+ and stripped from the tube. Photographic analysis of the tubes can quantify the degree of reduction (1; 5). The assessment of IRIS tubes by 1 found that when 30% of the ferrihydrite paint was removed from the tube, essentially all of the soils were under reducing conditions. The National Technical Committee for Hydric Soils (NTCHS) adopted the following as a technical standard to identify anaerobic conditions in soils: “Using IRIS‐tube data to verify anaerobic conditions requires at least three of five IRIS tubes to display iron removed from 30% of a zone 15 cm (6 in) long starting with 15 cm (6 in) of the soil surface” (9).
Alternate measurements of reduction in soils include the measurement of oxidation–reduction (Eh) potential using platinum (Pt)‐tipped electrodes and α,α dipyridyl dye. However, these methods reflect point measurements of conditions, whereas IRIS tubes offer a reflection of soil conditions over the duration they are installed. The major benefit that distinguishes IRIS tubes from other in situ methods that measure soil reduction is that they integrate a much larger area of the soil (i.e., versus measuring redox potential in a microsite associated with the Pt electrode tip). Because Fe is nearly ubiquitous in soil, IRIS tubes are a relevant measurement of reduction in soils nearly everywhere and play an important role in classifying hydric soils (5). In addition to displaying Fe3+ reduction, IRIS tubes can demonstrate sulfate (SO42−) reduction.
(3)
(4)
(6)
(7)The black‐colored FeS is unstable in aerobic conditions and thus will disappear when oxidized by O2 on exposure to air (15; 19). The deposition of FeS requires a specific set of conditions, which include organic matter, a SO42− and Fe3+ source, anaerobic microorganisms, and favorable reducing conditions (15). Because SO42− is a less favorable TEA than Fe3+, the deposition of FeS on IRIS tubes suggests that the reducing conditions are more severe in a soil displaying S reduction compared with only Fe3+ reduction. Although Fe3+ reduction is a valuable tool for hydric soil identification, SO42− reduction indicates stronger reducing conditions and should be evaluated for its use in identifying hydric soils. The striking contrast of black FeS precipitate on IRIS surfaces allows users to quickly and reliably identify reduction based on timely field observation. The objective of this study was to determine the minimum percentage of black staining (FeS) on IRIS tubes or panels indicative of strongly reducing conditions as corroborated by measured redox potential data and Fe removal from IRIS tubes.
MATERIALS AND METHODS
Soil was collected from the A horizons (0–25 cm) at Chorro Creek Ranch located at 35°19′49.03″ N and 120°45′17.97″ W in San Luis Obispo County, CA. The soil is classified as a fine‐loamy, mixed, superactive, thermic Pachic Haploxerolls [Haplic Phaeozems (Pachic)] found on the backslopes and summits of hills (4; 17). The parent material consists of alluvium weathered from sandstone and shale. The soil was well drained, with very high runoff and no ponding (18). The soil texture was determined using the hydrometer method as loam (49% sand, 30% silt, and 21% clay) (16). The bulk soil was passed through a 2‐mm‐diameter sieve to exclude coarse fragments. Organic C content measured via dry combustion was 2.2% and native S concentration measured by saturated extraction was 0.02% for the bulk homogenized soil (10).
IRIS Panel Construction
Iron oxide paint was synthesized following the procedure outlined by 12. Instead of using IRIS tubes, 30.5 cm by 30.5 cm by 0.3175 cm (12 in by 12 in by 1/8 in) polyvinylchloride panels were used. The panels were sanded with 220‐grit sandpaper, and a single coat of the IRIS paint was applied to the lower half of each side as an analog to the NTCHS criteria using IRIS tubes. Panels, as opposed to tubes, were used to obtain greater surface area and to increase ease and accuracy of analysis. For example, one double‐sided IRIS panel and one standard IRIS tube installed in soil 15 cm deep provided an observation area of 914 cm2 and 120 cm2, respectively.
Soil Mesocosms
Soil mesocosms were constructed in polypropylene bins (46 cm length by 35 cm width by 27 cm depth). Four IRIS panels were installed in each mesocosm and held in place with a wooden jig for the duration of the study. Four treatment groups of varying S levels were produced by adding S in the form of gypsum (CaSO4·2H2O) to obtain percent S‐added by mass of 0, 1, 2.5, and 5%. Three replicate mesocosms were created for each treatment. The 14 kg of air‐dry, sieved soil plus added gypsum were homogenized and added to the polypropylene bins in an even layer, resulting in a bulk density of 0.6 g cm‐3.
Municipal water was dechlorinated by setting water in open bins for 24 h to allow Cl2(g) to volatilize. A small sponge was placed in the corner of each mesocosm on the soil surface, and dechlorinated water was poured gently onto the sponge to minimize soil disturbance. A constant water depth of 5 cm above the soil surface was maintained throughout the duration of the study across all treatments. Soil mesocosms were kept at approximately 21°C and out of direct sunlight for the duration of the 11‐wk study. Treatments were randomly assigned within blocks.
Oxidation‐Reduction Potential and pH
Platinum‐tipped electrodes were constructed using the heat shrink method described by 24. The Pt‐tipped electrodes and a Ag/AgCl reference electrode connected to a high‐input resistance voltmeter (Accumet portable Ag/AgCl AP115 pH/ORP Meter, Fischer Scientific) were used to measure Eh at 15 cm below the soil surface to account for potential drift associated with the displacement of O2 as documented by 11 and 11. Five replicate Eh measurements were taken in each mesocosm on a weekly basis for 11 wk. Platinum‐tipped electrodes were installed and removed before and after each weekly measurement. Measured Eh values were averaged and corrected for the Ag/AgCl reference electrode by adding 200 mV. Because IRIS panels were removed from soil for analysis on Week 11, only Week 11 Eh values were used for comparison with IRIS panels. The probes were routinely checked to ensure proper function using Light's solution made up of 0.1 mol L−1 (NH4)2Fe2+(SO4)2·6H2O, 0.1 mol L−1 NH4Fe3+(SO4)2·12H2O, and 1 mol L−1 H2SO4 (7; 24). All electrodes fell within the Light's solution acceptable of range of 476 ± 20 mV. Prior to pH reading, the pH meter was calibrated using pH 4, 7, and 10 standard solutions. Soil pH was measured weekly by placing the pH electrode approximately 15 cm below the soil surface in each mesocosm.
IRIS Panel Analysis
After 11 wk of incubation, the IRIS panels were removed from the mesocosms and rinsed gently to remove any adhering soil. Photographs of both sides of the IRIS panels were analyzed for percent of FeS (black), Fe2+ removed (white or noncolored), and Fe3+ remaining on the panel (orange). Images of panels were cropped to display only the section of the panel that was installed in saturated soil (Fig. 1). The total number of pixels on each IRIS panel was determined using the histogram feature in Adobe Photoshop CS6. The area of Fe3+ oxide remaining was determined using the Selection, Color Range tool by selecting the range of colors characteristic of Fe3+ (red [R], green [G], and blue [B] ranges of 175–225, 100–125, and 20–80, respectively). The number of Fe3+ pixels was determined in the histogram feature. The same procedure was used to determine the number of FeS pixels, with the black FeS color range selected (R, G, and B ranges of 0–100, 0–100, and 0–100, respectively). The amount of Fe2+ removed was determined by subtracting the number of Fe3+ remaining and FeS pixels from the total number of pixels. The percentages of Fe3+ remaining, Fe2+ removed, and FeS precipitated were determined by dividing the number of pixels occupied by each color range by the total number of pixels. The percentages of Fe3+ remaining, Fe2+ removed, and FeS precipitated were analyzed as a one‐way ANOVA set in a randomized complete block design.
IRIS (Indicator of Reduction in Soil) panel displaying oxidized Fe (Fe3+, orange) remaining, reduced Fe (Fe2+, white or noncolored) removed, and FeS precipitate (FeS, black) cropped to the surface of the saturated soil. Dimensions are approximately 12 cm by 30 cm.
RESULTS AND DISCUSSION
Treatment versus Reduced S
Analysis of the IRIS panels allowed for the comparison of treatment (S concentration) to the percentage of FeS present on the panels (Fig. 2). Rapid photography of the IRIS plates was essential to ensure the FeS precipitate did not convert to hydrogen sulfide (H2S) gas before photographs were taken. Visible FeS precipitation increased as S concentrations increased. Treatment 0% S‐added had the lowest percentage of FeS precipitation, with a median of 0.2% black staining. Treatment 1% S‐added had a median of 2.5% FeS precipitation, whereas Treatment Groups 2.5 and 5% S‐added displayed medians of 5.2 and 7.7% FeS precipitation, respectively. The reduction of SO42− and subsequent formation of FeS on IRIS panels increased as S content increased. The percentage of FeS precipitation was significantly different between all treatment groups except between the 2.5 and 5% S‐added treatments (Fig. 2). Differences existed between the treatment groups with lower S added due to the increasing availability of SO42− to be used as a TEA by SO42−–reducing microorganisms. The availability of SO42− may have been sufficient in both the 2.5 and 5% S‐added treatment groups, resulting in the formation of similar percentages of FeS precipitation. Future studies may benefit from the incorporation of aqueous S to reduce microsite variability and greater S additions coupled with increased duration of saturation to examine the influence of high S contents.
Iron monosulfide (FeS) precipitation on IRIS (Indicator of Reduction in Soil) panels versus S treatment group. Box plots show median values (solid horizontal line), 50th percentile values (box outline), 90th percentile values (whiskers), and outlier values (circles, data falling outside 1.58 times the interquartile range above or below the upper or lower quartile). A one‐way ANOVA and Tukey's multiple comparison tests were used to determine differences between treatment groups. Boxplots of treatment groups with different letters are significantly different at the 0.05 level.
Redox Potential versus Reduced S and Fe
Throughout the duration of the 11‐wk study, Eh generally decreased (Fig. 3). Median Eh values for the 0, 1, 2.5, and 5% S treatment groups were 6.8, −6.5, −50.9, and −52 mV, respectively, when the IRIS panels were removed at Week 11. Based on the technical standard for anaerobic conditions, soil Eh measurements less than 205 mV at pH 6.5 meet the requirements for hydric soils (slope of the technical standard line is Eh = 595–60 pH) (9). All Eh measurements collected when IRIS panels were removed (Week 11) were lower (between 125 and 273 mV less) than the technical standard based on measured pH and therefore met the anaerobic conditions criteria for hydric soils (Fig. 4).
Oxidation‐reduction potential (Eh) measured for the 11‐wk duration of the study. Box plots show median values of five Pt‐tipped electrodes in three replicate mesocosms for each treatment group (solid horizontal line), 50th percentile values (box outline), 90th percentile values (whiskers), and outlier values (circles, data falling outside 1.58 times the interquartile range above or below the upper or lower quartile).
Comparison of oxidation reduction potential (Eh) and the mean percent of FeS precipitate present on and Fe removal from IRIS (Indicator of Reduction in Soil) panels (p = 0.001). Based on the technical standard for anaerobic conditions, soil Eh measurements less than 205 mV at pH 6.5 (mean soil pH measured in mesocosms) meet the requirements for hydric soils (Eh = 595– 60 pH) (9). Therefore, all panels exhibited reduced soil environments.
The FeS precipitation was compared with the measured Eh at the termination of the study. As Eh decreased, the percentage FeS precipitation on the panels increased (Fig. 4). Due to the low thermodynamic favorability of SO42− as a TEA, lower Eh values corresponded with greater SO42− reduction. Given the heterogeneity of soil and the variability within Pt‐tipped electrode readings described by 14, the R2 value of 0.69 indicated a significant correlation (p ≤ 0.001) between Eh and percentage of FeS precipitation on IRIS panels. The moderate negative correlation provided evidence that SO42− was increasingly reduced to FeS in more severe anaerobic conditions.
The percentage of Fe removal from IRIS panels was also compared with the Eh measured at the termination of the study (Fig. 4). All mesocosms met the NTCHS criteria for the technical standard for anaerobic conditions, but there is no significant relationship between Eh and Fe removal from the panels. There also appears to be no conclusive influence of treatment group on the amount of Fe removed from the IRIS panels. Medians of 42, 27, 36, and 30% Fe removal were measured on panels removed from the 0, 1, 2.5, and 5% treatment groups, respectively.
Interaction between the reduction of Fe3+ and SO42− likely resulted in a decrease in Fe removal observed on IRIS panels in the mesocosms with added S. A comprehensive study of these interactions could improve understanding of the mechanisms involved and of the degree of interaction. At low redox potentials (<−200 mV) and near neutral pH, the stability lines for Fe3+:Fe2+ and SO42−:H2S intersect, with SO42–reduction becoming increasingly thermodynamically favorable. This is likely the reason for the increase in FeS precipitation because Eh decreases, whereas Fe removal (Fe3+ reduction) stagnates.
Reduced SO42‐ versus Reduced Fe3+
The established technical standard for Fe reduction on IRIS tubes is 30% Fe oxide paint removal from three of five IRIS tubes (9). The percentage of FeS precipitation was compared with the amount of Fe removal from the IRIS panels. There was a significant positive correlation (R2 = 0.39, 0.47, 0.48, and 0.13 for treatment group 0, 1, 2.5, and 5%, respectively; p < 0.009) between FeS precipitation and Fe reduction on IRIS panels (Fig. 5). Each of the IRIS panels meeting the 30% Fe removal requirement by the NTCHS displayed ≥9% FeS precipitation (Fig. 5). This conservative value would fail to capture the majority of panels also installed in anaerobic conditions (based on Eh measurements) (Fig. 5, dashed horizontal line). When IRIS panels had ≥2% FeS precipitation, only 77% of the panels (49/64) met the NTCHS criteria for reduction using IRIS panels, whereas 100% of these panels (64/64) met the NTCHS criteria for anaerobic conditions using measured Eh (Fig. 4 and 5).
Comparison of percent FeS precipitation and Fe removal on Indicator of Reduction in Soil (IRIS) panels. Lines represent the current standard for Fe removal on IRIS tubes (30%) and the suggested level of FeS precipitation to indicate anaerobic conditions (2%).
Within the treatment groups that had added S above native concentration, interactions between SO42−–reducing microorganisms and Fe3+–reducing microorganisms likely resulted in lower Fe3+ reduction observed on the IRIS panels relative to the 0% S treatment group. Beneficial conditions for both Fe3+ and SO42− reduction (Eq. [6] and [7]) were present in the mesocosms with adequate organic C, favorable temperature, and anaerobic conditions. This may have resulted in less evidence of Fe3+ reduction than would be expected given the low redox potentials measured. Future work should address the potential interactions between the processes involved in Fe3+ reduction and SO42− reduction.
CONCLUSION
The objective of this study was to determine the minimum percentage of black FeS staining on IRIS tubes or panels indicative of reduction as corroborated by measured Eh data and Fe reduction from IRIS tubes. Visual SO42− reduction on IRIS panels was related to Fe3+ reduction using IRIS tubes and measured redox potential using Pt‐tipped electrodes. In all cases, when FeS staining was ≥9%, Fe3+ reduction exceeded the existing technical standard of 30% removal. Comparison of FeS to measured Eh values confirmed that FeS staining was increasingly present under more severe reducing conditions. Precipitation of FeS on IRIS panels increased with higher S concentrations. Using 2% FeS precipitation on the IRIS panels as an indicator of an anaerobic soil environment will allow practitioners and researchers to quickly visually analyze IRIS tubes and panels in the field.
The observance of FeS precipitation on IRIS tubes and panels is useful for the identification of hydric soils. The results of this research demonstrate that observation of SO42− reduction on IRIS surfaces is a viable method for determining hydric soil status and ultimately performing wetland identification and delineations. Evidence of SO42− reduction (FeS precipitation) on IRIS surfaces proves highly reducing conditions and could be used in lieu of the requirements for Fe removal from IRIS surfaces. Future experiments should explore inundating the soil for longer and shorter durations, varying organic C type and amount, field verification, and investigating the interaction between Fe3+ reduction and SO42− reduction to see how these parameters influence the soil redox environment.
ACKNOWLEDGMENTS
Partial funding for this research was provided by the California State University Agricultural Research Institute (ARI). The authors thank Craig Stubler and Mateo Echabarne (California Polytechnic State University) for field and lab assistance, Robert Vaughan (Red Castle Resources Inc.) for technical assistance and scholarly discussion, and two anonymous reviewers for insightful and critical comments that helped to improve this manuscript considerably.
