Published online 2 December 2005
Published in Soil Sci Soc Am J 70:183-191 (2006)
DOI: 10.2136/sssaj2004.0323
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Pedology and Wetland Soils
Development and Evaluation of Iron-Coated Tubes that Indicate Reduction in Soils
B. J. Jenkinson and
D. P. Franzmeier*
Dep. of Agronomy, Purdue Univ., Lilly Hall of Life Sciences, 915 W. State St., West Lafayette, IN 47907-2054
* Corresponding author (dfranzme{at}purdue.edu)
 |
ABSTRACT
|
|---|
Soil drainage conditions are important to land use decisions. Traditionally, anaerobic conditions induced by poor drainage have been evaluated by observing soil color related to Fe and Mn oxides, using 
, -dipyridyl dye, measuring dissolved O2, and measuring EH. We believe that there is further need for a device that is scientifically sound and easy to use. Therefore, our goals were to develop and test a device that mimics natural soil processes, visually indicates soil reduction, and is robust. Our concept was to coat a rod or tube with a colored soil mineral that dissolves on reduction, insert the device into a soil, remove it after a few weeks or longer, and observe if some of the coating had been lost. If the coating was not dissolved, no reduction occurred, but if it was dissolved, reducing conditions must have prevailed. After trying several kinds of coatings and tubes, we chose ferrihydrite (FH) coating on polyvinyl chloride (PVC) pipe. We call the device an Indicator of Reduction in Soils (IRIS). As the study progressed we added semi-quantitative interpretations by measuring depleted areas using a digital camera and image analysis. We tested IRIS tubes in the lab and in soils in Indiana, Minnesota, and North Dakota, and concluded they performed as expected. Reduction rates increased between February and April and were related to increasing soil temperature, turnover (flux) of soil OC, and content (inventory) of OC. Reduction rates decreased after April, presumably because the nutrient supply for microbes decreased.
Abbreviations: Ac, area of FH coating in contact with the soil • Ad, area from which some FH had been depleted • D, percentage of Ac from which some FH had been depleted • DO, dissolved O2 concentration • FH, ferrihydrite • IRIS, indicator(s) of reduction in soils • OC, organic carbon • PVC, polyvinyl chloride • UDD, upper depth of FH depletion
|
INTRODUCTION
|
|---|
THE NATURAL DRAINAGE or oxidation–reduction status of soils is important for understanding ecosystems, preserving wetlands, installing tile drainage for crop production, determining the suitability of a soil for constructing houses and other structures, and designing onsite wastewater disposal systems. The oxidation–reduction status of soils is reflected in soil color. The reddish, brownish, and yellowish color of soils is due mainly to Fe oxide minerals, a term that includes oxides, oxyhydroxides, and hydrated oxides (Schwertmann and Taylor, 1989). Common Fe oxide minerals in soils and their colors are goethite, yellowish-brown; hematite, bright red; lepidocrocite, orange; and FH, reddish-brown (Bigham et al., 2002). Soil color resulting from Fe oxides is an important property and is used in soil classification systems worldwide. Ferrihydrite was called "amorphous Fe oxide" before it was learned the material actually has short-range ordering of atoms. It usually has a greater surface area than goethite and hematite (Cornell and Schwertmann, 2003).
Even at low concentrations in soils, Fe oxides have a high pigmenting power. Electron microscopy shows that small discrete Fe oxide particles occur on the surface of much larger silicate clay particles rather than forming a paint-like coating (Bigham et al., 2002). The underlying silicate minerals are light gray, as shown by a soil sample from which Fe oxides have been extracted with a reducing agent. Thus, the relationship of Fe oxide to silicate clay is more like sesame seeds on a bun than glaze on a donut. Our subsequent use of "coating," however, does not imply a paint-like covering.
Soil microorganisms produce electrons during metabolic oxidation of organic matter, and these electrons must be transferred to an electron acceptor. In freely drained soils, the acceptor is O2 (Schwertmann and Taylor, 1989). Oxygen diffuses through air much faster than through water, so when a soil is saturated with water, O2 usually becomes deficient. Electrons are then accepted, in order, by nitrate, Mn, and Fe. When Fe3+ in Fe oxide is reduced, the oxide dissolves and Fe2+ goes into solution. Then the grayish color of the underlying silicate minerals and quartz appears. The Fe2+ in solution may move only a short distance within a horizon, and thus be available for later oxidation as Fe oxide, or it may move out of the soil horizon entirely. According to equilibrium reactions amorphous Fe oxide (FH) is reduced at a lower redox potential (EH) than hematite and goethite (Lindsay, 1979), suggesting that as a soil becomes anaerobic, FH is reduced and dissolved before goethite and hematite. This was confirmed in the laboratory by Schwertmann and Taylor (1989).
Four soil conditions are needed to reduce Fe in soils: saturation with stagnant water, presence of microorganisms, a supply of organic C (OC), which serves as an energy source for these organisms, and suitable soil temperatures (Bouma, 1983; Jenny, 1980; Vepraskas and Faulkner, 2001). Exactly which temperatures are suitable is open to debate. Evans and Franzmeier (1988) showed that soil reduction features were much better correlated with the time the soil was saturated when soil temperatures were >5°C than with the time the soil was saturated when temperatures were 
5°C. Also, Meek et al. (1968) found more Fe2+ and Mn2+ in soil solutions a few day after irrigation when the soil temperature was 
35°C than when it was 19°C. On the other hand, Jenkinson et al. (2002) reported that EH decreased in soil horizons that were not entirely saturated and the soil temperature was between 0 and 5°C. They postulated that some reduction occurs in saturated microsites of these horizons when they are cold. Also, Megonigal et al. (1996) concluded that, in southeastern USA where soil temperatures ranged from 5 to 30°C, saturation might cause anaerobic conditions to develop as rapidly in winter as in summer. From these studies it appears that microbial reduction is much slower, but not negligible, at temperatures <5°C than at warmer temperatures, but that >5°C microbial activity may or may not depend on soil temperature.
Currently, oxidation–reduction characteristics are inferred from soil color, measured with platinum (Pt) electrodes (Bartlett and James, 1993), identified by red color produced using , ' dipyridyl dye (Childs, 1981), and inferred from rusting of Fe nails (Owens, 2001). In addition to these methods, we believe that there is a need for a device that visually indicates current soil reduction processes, mimics natural soil processes, is easy to use, and is robust. Therefore, the main objectives of the study were to develop and test such a device. Our idea was to coat a rod or tube with Fe oxide, place the device in the soil for some time period, remove it, and observe in the field whether or not some Fe oxide had been dissolved. Ideally, the processes affecting the device should mimic soil processes. As the work progressed, we add some semi-quantitative estimates of the degree of FH removal. This paper describes how we developed and tested the device, which we called an Indicator of Reduction In Soils (IRIS), and how the results relate to soil environmental conditions, mainly temperature and soil OC content.
|
MATERIALS AND METHODS
|
|---|
Our general concept was to bond an Fe oxide mineral subject to reduction on a rod or tube, with the idea that the tube could be easily installed and removed from the soil and visually assessed for the effects of reduction by the depletion of the oxide coating. If the colored oxide material occurred naturally in soils, the device would closely represent soil processes. We tried several combinations of Fe oxide minerals (FH, lepidocrocite, and goethite) applied to wooden dowel rods and tubes and rods of various plastic materials (polyvinyl chloride, nylon, acetal homopolymer, and porous polyethylene). For a detailed description of the materials and methods utilized in the development of IRIS, refer to Jenkinson (2002). We chose FH as the oxide because it is more readily reduced than the other oxides and because it adheres better to the tube surface. We chose white polyvinyl chloride (PVC) tubing because FH adheres to it well, it shows good color contrast with the FH, is inexpensive, and is readily available.
Construction of IRIS
The methods for preparing IRIS evolved over time by trial and error (Jenkinson, 2002). The methods described are those used for IRIS Numbers 13 to 77 (Table 1). Construction methods for IRIS 1 to 12 were somewhat different. We believe that despite this difference in manufacturing, the generalized responses were similar enough to present in this study, but further control studies are needed.
View this table:
[in this window]
[in a new window]
|
Table 1. IRIS number; location, soil series and natural drainage class; dates and water table depths when installed; time IRIS were in the soil; upper depletion depth, area of IRIS in contact with the soil, percentage of ferrihydrite (FH) coating depleted, and rate of depletion.
|
|
Ferrihydrite was prepared by dissolving FeCl3 in water and adding KOH to the solution according to Schwertmann et al. (1995). The FH suspension was dialyzed in molecular-porous membrane tubing for 7 d in a continuous flow bath that had a fresh water exchange rate of 50 mL h–1lot–1 of FH to ensure purity. Precipitated FH was stored as a suspension until it was applied to the tubes. In some cases, the FH suspension was stored for several months before application, but this did not seem to compromise its efficacy as a reduction matrix.
White PVC pipe with an outside diameter of 1.9 cm was cut to 60 cm lengths and cleaned with acetone. This pipe is readily available from hardware and plumbing suppliers where it is sold as "1/2 inch id Schedule 40 PVC pipe." Over time it was learned that FH would adhere better if the pipes were first placed in a lathe and sanded with 100-grit sandpaper.
Initially, FH suspension was applied to the PVC pipe segments with a foam brush by brushing lengthwise along the pipe. Later it was applied by moving the brush along the pipe while the pipe was turning in a lathe to give a circumferential application pattern. The coating was dried in a stream of hot air (100–480°C) using an air-flow heat gun for 45 s. A second coat was applied in a similar fashion after the initial coating had completely dried. The circumferential coating pattern facilitated identification of abrasion marks caused by installation or removal of the tubes from the soil.
Other methods include water table depth by observation wells and piezometers, soil temperature by thermocouples, EH by Pt electrodes, and OC by the Walkley–Black method (Soil Survey Staff, 1996). Laboratory methods are described in more detail in Jenkinson et al. (2002) and Jenkinson (1998).
Laboratory Study
IRIS were installed in soil samples in two 20-L buckets in the laboratory under conditions that we believe were ideal for reduction. First we installed a small standpipe outside of the bucket to measure water depth within the bucket. Then 3 L of sand were poured into the bottom of each bucket, the sand was covered with geofabric, and 15 L of the A horizon of the Clermont soil from the Muscatatuck site were added. Five Pt electrodes were installed in the soil around the side of the bucket and a KCl electrode was installed in the center to measure EH (Faulkner and Patrick, 1991). The electrodes were connected to a data logger to record readings hourly for 25 d. Four IRIS tubes, 25.5 cm long, were also installed in each bucket. Water with 25 g L–1 sucrose was added to the bucket to saturate the soil to the surface and to supply food for the microbes. Water was also added periodically to maintain the water level at the soil surface. Soil pH was measured with colored indicators brom thymol blue, brom cresol green, and chlor phenol red at 2- to 3-d intervals, and dissolved O2 was measured using Chemets color indicators (Fisher Scientific) and a Thermo Orion 810+ dissolved O2 probe (Jenkinson, 2002).
Installing and Retrieving IRIS in the Field
Study sites were in Indiana, Minnesota, and North Dakota (Fig. 1 , Table 1). All except the Eagle Creek sites (W. Hosteter, personal communication, 2001) were part of the Natural Resources Conservation Service Wet Soil Monitoring program. Soil characterization data are available for many of the pedons. Classification and pedon ID of the pedons sampled are in Table 2. Not sampled are Brookston (Typic Argiaquoll) and Giese (Mollic Epiaquept). Characterization data and more detailed classifications are given in Jenkinson (1998) for Indiana soils, Hopkins (1996) for North Dakota soils, and Feigum (2000) for Minnesota soils.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Map showing locations of study areas. 1. Muscatatuck Wildlife Refuge, Jennings County, IN. Lat. 38° 56' 46'', Long. 85° 47' 27''. Moore's Woods, Parke County, IN. Lat. 39° 56' 48'', Long. 87° 7' 52''. Shades State Park, Parke County, IN. Lat. 39° 55' 52'', Long. 87° 5' 39''. Jasper Pulaski Fish and Wildlife Area, Jasper County, IN. Lat. 41° 9' 12'', Long. 86° 59' 7''. 5. Eagle Creek Nature Preserve, Marion County, IN. Lat. 39° 51' 56'', Long. 86° 16' 56''. 6. Sheyenne National Grassland, Ransom County ND. Lat. 46° 26' 36'', Long. 97° 24' 20''. 7. Hamden Slough National Wildlife Refuge, Becker County, MN. Lat. 47° 0' 25'', Long. 95° 59' 14''. 8. Superior Lobe Till, St. Louis County, MN. Lat. 46° 56' 14'', Long. 92° 1' 15''.
|
|
In the field, we bored a 1.9-cm diam. hole with a soil probe or auger to a depth of 50 cm. In moist soil, we simply pushed the IRIS into the hole until it was seated at 50 cm. If the soil was dry, we poured water into the hole and installed IRIS after water had soaked into the soil and softened it. This lessens the possibility of removing FH by friction during installation. The upper 10 cm of the IRIS remained above the soil surface to facilitate recovery and identification. This method resulted in good contact between the soil and the tube. All IRIS installed outdoors except Numbers 19 and 69 in Table 1 were installed in pairs. In the lab, two sets of four replications were installed (70–77). IRIS Numbers 1 to 12 represent only one tube of the pair because no FH was removed from either tube. Results from both tubes of a pair are reported for IRIS numbers 13 through 68 (e.g., set 1a and 1b are replicates). IRIS tubes 70 to77 were installed in replicates of four in the laboratory study. To remove the IRIS, we moved it back and forth in the hole to loosen it. Then it could be removed by hand or by gripping it with pliers. If it did not loosen, we inserted a spade into the soil about 10 cm back from the IRIS. Rocking the spade back and forth, and then rocking the IRIS loosened it so it was easy to pull out.
Estimating Loss of Ferrihydrite from Surface of Tubes
The appearance of the IRIS surface can be evaluated qualitatively by determining whether or not some FH was removed, or it can be evaluated semi-quantitatively as described here. After we removed an IRIS from the soil, we rinsed it with water, dried it, and photographed it with a digital camera. We took a photo, rotated the tube 120°, took another photo, and repeated the process two times to obtain three photos representing 120° rotations. Thus, the entire surface of the IRIS was photographed, but with some distortion. We combined the three images into a composite using Adobe Photoshop software. Using that software, we tentatively delineated areas on the image that appeared to show some FH depletion and then inspected the actual tube to confirm that FH had been depleted from that area. In most cases, inspection of the tube confirmed FH depletion and we accepted the tentative delineation. In a few cases, however, the tentative delineation represented areas that might not have been coated as heavily originally, might have been scratched when the tube was removed from the soil, or might represent an anomaly in the image, and we rejected the tentative delineation.
Next, we measured the area of each polygon and added them to give the total depleted area, Ad (mm2). Then, we calculated the area of the tube in contact with the soil, Ac (mm2). From these two measurements we calculated the percentage of the area of the tube in contact with the soil from which some FH had been depleted, D (D = Ad/Ac x 100). We also calculated the rate of depletion, Dr, by dividing D by the number of days the IRIS was in the soil.
In future work, images could be evaluated more quantitatively by sorting pixels into different color classes so the degree of FH removal could be estimated. For example, the original coating was usually yellowish red, approximately 5YR 5/8 Munsell color, but it varied somewhat. The bare PVC pipe was white (N8/0). Various degrees of removal within that color range could be estimated. This kind of analysis, however, will require more uniform application of the FH coating to ensure that the color differences are due to removal rather than application. In future work, manufacturing technology and image analysis must progress in tandem.
|
RESULTS
|
|---|
Overview
The following discussion is based on the assessment of FH coating removed from the IRIS surface. As a point of reference, the surface of an IRIS never installed in a soil is shown in Fig. 2A . The IRIS changed little during the 14 d the IRIS was in a soil that was moist but never saturated (Fig. 2B). The difference in color between 2A and 2B is due more to differences in the original color of the coating than to changes that occurred while the tube was in the soil. When installed in horizons that were saturated and became anaerobic, IRIS lost some (Fig. 2C) to all of the FH coating (Fig. 2D). Occasionally, sections of IRIS that were never below the water table for extended periods lost some FH. This partial removal in the absence of extensive anaerobic conditions can be explained by two mechanisms. In some cases, an organic mass became attached to an IRIS (Fig. 2E), and if this mass was removed, some FH was removed with it. This is probably due to localized chelation (Jenny, 1980). A second mechanism of non-anaerobic FH removal is microsite reduction in an otherwise aerobic environment (Fig. 2F). Usually these anomalous areas were found on only one of the duplicate installations. Therefore it is best to examine duplicate IRIS to estimate the depth of FH depletion and the anaerobic status of the soil.
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 2. Color photos of IRIS. The IRIS numbers in parentheses refer to Table 1. (A) A new IRIS never installed in a soil (not in Table 1), (B) IRIS installed in a soil for 14 d showing no removal of FH (46), (C) IRIS installed in the soil for 14 d showing partial removal of FH from part of the tube (48), (D) IRIS showing complete removal of FH from all of the tube (61), (E) IRIS showing some FH removed by organic mass attached to tube surface (7), (F) IRIS showing FH removed at a microsite above water table (9).
|
|
We noticed several interesting patterns on the IRIS surface, speculated on how they were formed, and showed them to colleagues who are specialists in other disciplines for their comments. We observed donut-shaped depletion patterns on several IRIS tubes (Fig. 3A
). D. Huber (personal communication, 2004), a microbiologist, said they look like patterns formed by bacterial cultures in Petri dishes. Linear depletion patterns (Fig. 3B, 3C) probably reflect IRIS tube proximity to an organic matter source like roots. We also noticed that some areas were darker brown than the original coating (Fig. 3D), and interpreted that to represent deposition of Fe oxide on the tube while it was in the soil. D. Schulze (personal communication, 2004), a mineralogist, examined some of the dark areas with a synchrotron X-ray microprobe and found that they also contained Mn, which confirms deposition from the soil because no Mn was applied to the tube. These patterns need further study and suggest additional uses for IRIS tubes.
View larger version (74K):
[in this window]
[in a new window]
|
Fig. 3. Color photos of IRIS. The IRIS numbers in parentheses refer to Table 1. (A) IRIS showing FH removal in a "white donut" pattern (42), (B and C) IRIS showing removal in a linear pattern, probably along a root (25, 32), (D) IRIS showing secondary deposition on right center of tube (66).
|
|
Laboratory Study
We created conditions in the laboratory that we thought were ideal for reduction processes. Immediately after saturation, both dissolved O2 concentration (DO) and EH decreased rapidly as seen in photos (Fig. 4
) and graphs (Fig. 5
). We believe that aerobic and facultative microorganisms very quickly reduced O2 and probably nitrate driving down EH and creating suitable conditions for obligate anaerobic microorganisms to start metabolizing C and reduce FH. By Day 2, DO had dropped to 2 mg L–1 and EH dropped to 
300 mV, below the Fe oxidation–reduction line (oxidizing-reducing line, Fig. 7.6, McBride, 1994). Between Days 2 and 7, DO leveled off, and by Day 9 it had dropped to almost zero. After Day 2, EH continued to fall until Day 6; remained low until Day 8, and then increased. On Day 8, Fe3+ began to precipitate on the surface of the saturated soil and on IRIS, wires, and glass tubes above the soil surface. The EH increased because the electrodes respond to the concentrations of both Fe species, and the system was losing Fe2+ due to oxidation and precipitation at a greater rate than microbial respiration was reducing Fe3+.
View larger version (78K):
[in this window]
[in a new window]
|
Fig. 4. Color photos of IRIS. The IRIS numbers in parentheses refer to Table 1. The dotted black line represents the soil surface. (A) IRIS removed after 15 d in laboratory study (77), (B) IRIS removed after 24 d in laboratory study (73), (C and D) Duplicate IRIS show depletion of FH below 2 cm (53, 54); 4C shows several "white donut" patterns around 14 and 30 cm; 5D shows diagonal linear reduction pattern at 36 cm, probably caused by a root. (E) IRIS removed from natural soil during the year 2000 (19), (F) IRIS removed from filled soil pit near E the same year, (21), (G) IRIS removed from natural soil during 2001 (43), (H and I) (35 and 36) are duplicate IRIS, Fig. 5H shows depletion of FH below 2 cm, Fig. 5I represents three photos at 120° rotation and also shows depletion below 2 cm and shows "white donuts" at 14 cm and 18 cm, (J) IRIS installed in a Hamar soil shows complete removal of FH. (61), (K) IRIS installed in a Winger soil shows almost complete removal of FH from the upper part of the tube (65).
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 5. Changes in pH, dissolved O2 concentration, and redox potential (EH) during laboratory study. The Fe oxidation–reduction line represents the EH for Fe reduction at the measured pH (Fig. 7.6, McBride, 1994).
|
|
As EH decreases, soil pH increases and Fe3+ is reduced as predicted from the reaction:
 | [1] |
The pH change lags behind the EH change, however. EH reached a minimum on Day 6 and pH reached a maximum on Day 11.
Four IRIS were removed from the soil after 15 d and four after 24 d (Fig. 4A, 4B). IRIS removed after 15 d (no. 75–78) lost an average of 12% of their FH coating (Fig. 4A), but those removed after 24 d lost an average of 76% of their FH (Fig. 4B). The rate of removal during the first 15 d was 112 mm2 d–1, and during the next 9 d was 800 mm2 d–1. Even though reduction of Fe theoretically began after Day 2, most reduction of FH took place between Days 15 and 24 as illustrated in Fig. 4. Apparently the microbial reduction of Fe takes several days to gain momentum. The maximum rate of depletion in the lab was slower than some rates in the Washtenaw soil (IRIS no. 28–30), however.
The dark reddish band on the tubes near the soil surface grew significantly between Day 15 (Fig. 4A) and Day 24 (Fig. 4B). We believe the band formed when FH on tubes and soil Fe-oxide minerals were reduced to soluble Fe2+ below the soil surface, Fe2+ diffused upward in the soil, rose up with the water on the PVC pipe by adhesion and cohesion, and was oxidized and reprecipitated to form the band where atmospheric O2 was plentiful.
Field Studies
We installed IRIS in several soils (Table 1, IRIS 1 to 23) in late 1999 and early 2000. In previous years the water tables in all these soils rose to near or above the soil surface (Jenkinson, 1998, Jenkinson et al., 2002). During the winter and spring of 2000, however, the water table did not rise to the lower depth of the IRIS in several installations (IRIS 1 to 12, Table 1), and the tubes changed very little in appearance while they were in the soil (Fig. 2B). This shows that little of the FH coating was removed when there were no reducing conditions in the soil.
In most installations, there was a fairly definite line on the IRIS below which evidence of reduction increased significantly in both IRIS in a set, for example, below 2 cm in Fig. 4C and 4D. We called this line the Upper Depletion Depth (UDD), the depth below which evidence of reduction increases significantly in both tubes of an IRIS set. It is fairly distinct in some installations (Fig. 4K), probably where the water table level was relatively uniform during the installation time, but less distinct in other installations (Fig. 4C). The line was easier to determine when examining the IRIS themselves rather than the photos of IRIS. Above the UDD line there were occasional depleted zones that we attributed to short periods of saturation, microsites of reduction in an otherwise aerobic horizon, or chelation. Usually these anomalous zones occurred on only one tube of a set. Below this line the amount of FH removal varied from some (Fig. 2C) to complete removal (Fig. 2D). In the field, as in the lab studies described above, FH removal was related to low EH measurements (Jenkinson, 1998, Jenkinson et al., 2002; Hopkins, 1996; Feigum, 2000) and, in spot checks, to development of red color with , ' dipyridyl dye (Childs, 1981).
Figure 4 compares two IRIS installed in the natural Avonburg soil (Fig. 4E, 4G) with one installed in the nearby filled pit (Fig. 4F) in which the soil had been described and sampled (Jenkinson, 2002). The water table in the natural soil fluctuated from near the surface to >2.5 m deep, but in the pit the water table was almost always near the surface. Some FH was removed from IRIS in the natural soil, but practically all FH was removed from the one in the pit. We attribute greater FH removal to the longer time of saturation in the upper 50 cm and to mixing of organic matter (OM) from leaves and severed roots in the fill material.
Figures 4H and 4I show two IRIS tubes installed in the Clermont soil. Three views of one tube (Fig. 4I) are shown, illustrating the 120° rotation used for analyzing the images. Depletion was patchy, and apparently Fe (and probably Mn) was reprecipitated in some areas (14 cm depth, Fig. 4I).
Figures 4J and 4K represent the Hamar and Winger soils of North Dakota and Minnesota, respectively. Almost all of the FH was depleted from all (Fig. 4J) or part (Fig. 4K) of the IRIS.
Rate of Reduction
Figure 6
represents data for several IRIS pairs (depicted as bars in the individual graphs) installed in four soils from February through May 2001. The beginning of the bar represents installation. The end represents removal, and the height of the bar above the x-axis represents the rate of depletion, Dr. The graphs also show water table depth, EH if determined, soil or air temperature, and the EH at which Fe3+ is reduced in soil (McBride, 1994).
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 6. Comparison of water table (WT) depth, redox potential (EH), temperature (T) of soil at 50-cm depth or of air, and dissolution rate for four soils: (A) Clermont (5 g C kg–1 soil); (B) Avonburg (7 g kg–1); (C) Washtenaw (17 g kg–1); and (D) Brookston (26 g kg–1). The organic C values represent the weighted average for the upper 50 cm of soil. The horizontal dashed line represents the theoretical EH for Fe reduction at the pH of the soil characterization sample (Fig. 7.6, McBride, 1994).
|
|
In all four soils the depletion rate tended to increase between February and April as soil temperature increased. After April, the rate decreased, as temperature continued to increase. We hypothesize that the increasing depletion trend represents a microbial response to increasing temperature and availability of OC. All of the Indiana sites are in forest. We observed that the layer of fallen leaves in the autumn was a few centimeters thick. During the winter, the layer was somewhat compressed, but it mainly remained intact. In early spring, however, the rate of leaf decomposition increased, and by mid summer, there were very few leaves on the mineral soil surface. We assume that during the period of rapid leaf decomposition, much soluble OC and other nutrients were released to the soil solution and leached downward. Perhaps equally important, is the OC from roots that die in place each year (Jenny, 1980). The activity of Fe-reducing and other microbes probably increased with warming temperature and increasing availability of nutrients, which caused the increase in reduction rate from February to April. By the end of April, however, microbes and the flush of new tree growth had utilized most of these nutrients, and the rate of Fe reduction decreased. We cannot separate the individual effects of temperature and nutrients in this process, however.
In addition to this annual flux of OC, the OC content of the soil (inventory) is also related to rate of reduction. The rate of FH removal (500–1000 mm2 d–1) is much greater in the Washtenaw and Brookston soils which have 17 and 26 g OC kg–1 soil, respectively, than in the Clermont and Avonburg soils (200 mm2 d–1), which have 5 and 7 g OC kg–1 soil (Fig. 6). Apparently, in these wet soils the low soil OC content limits the rate of reduction. The relative impact on microbial activity of OC flux and OC inventory needs further study.
|
CONCLUSIONS
|
|---|
At the onset of the study we had the idea of developing a device that could be placed in the soil for a period of time, removed, and observed in the field to learn if the soil had been anaerobic during the time it was in the soil. The device would consist of an inert rod or tube coated with a naturally occurring colored oxide that could be reduced and dissolved. For scientific and practical reasons, we chose PVC pipe as the inert tube, and FH, a reddish-brown Fe-oxide mineral, as the indicator coating. We called the device an Indicator of Reduction in Soils (IRIS). As the project developed, we made some semi-quantitative estimates of the degree of FH loss. No FH was dissolved when IRIS were installed in soils that were moist but not saturated. Both in the lab and the field, some reduction was observed after 2 wk, but significant removal required longer time periods. Under some field conditions, all FH was removed. Reducing conditions during the time of IRIS installation were confirmed with Pt electrodes and , ' dipyridyl dye. Thus, IRIS have the potential to help identify hydric soils. During late winter and early spring, FH reduction rates increased with increasing soil temperature and presumed greater availability of OC and other nutrients. Also, soils with medium and high OC contents had faster reduction rates than those with low OC content. IRIS have potential for studying redox processes such as microbial processes and redeposition of Fe and Mn. They could be analyzed more quantitatively with more uniform application of FH coupled with more rigorous image analysis.
|
ACKNOWLEDGMENTS
|
|---|
This work was supported in part by a grant from the USDA NRCS Wet Soil Monitoring Program administered by W.C. Lynn.
Received for publication October 1, 2004.
|
REFERENCES
|
|---|
- Bartlett, R.J., and B.R. James. 1993. Redox chemistry in soils. Academic Press, New York.
- Bigham, J.M., R.W. Fitzpatrick, and D.G. Schulze. 2002. Iron oxides. p. 323–366. In J.B. Dixon and D.G. Schulze (ed.) Soil mineralogy with environmental applications. SSSA Book Series No. 7. SSSA, Madison, WI.
- Bouma, J. 1983. Hydrology and soil genesis of soils with aquic moisture regimes. p. 253–281. In. L.P. Wilding et al. (ed.) Pedogenesis and soil taxonomy, I. Concepts and interactions. Elsevier, Amsterdam, The Netherlands.
- Childs, C.W. 1981. Field test for ferrous iron and ferric-organic complexes (on exchange sites or in water-soluble forms) in soils. Aust. J. Soil Resour. 19:175–180.[CrossRef]
- Cornell, R.M., and U. Schwertmann. 2003. The iron oxides: Structure, properties, reactions, occurrences, and uses. 2nd ed. Wiley-VCF, Weinheim, Germany.
- Evans, C.V., and D.P. Franzmeier. 1988. Color index values to represent wetness and aeration in some Indiana soils. Geoderma 41:353–368.
- Faulkner, S.P., and W.H.J. Patrick. 1991. Redox processes and diagnostic wetland soil indicators in bottomland hardwood forests. Soil Sci. Soc. Am. J. 56:856–865.
- Feigum, C.D. 2000. Evaluation of a drained lake basin and catchment utilizing soil, landscape position, and hydrology. Ph.D. diss. North Dakota State Univ., Fargo.
- Hopkins, D.G. 1996. Hydrologic and abiotic constraints on soil genesis and natural vegetation patterns in the Sand Hills of North Dakota. Ph.D. diss. North Dakota State Univ., Fargo.
- Jenkinson, B.J. 1998. Wet soil monitoring project on two till plains in south and west-central Indiana. M.S. thesis. Purdue Univ., W. Lafayette, IN.
- Jenkinson, B.J. 2002. Hydrology of sandy soils in northwest Indiana and iron oxide indicators to identify hydric soils. Ph.D. thesis. Purdue Univ., W. Lafayette, IN.
- Jenkinson, B.J., D.P. Franzmeier, and W.C. Lynn. 2002. Soil hydrology on an end moraine and a dissected till plain in west-central Indiana. Soil Sci. Soc. Am. J. 66:1367–1376.[Abstract/Free Full Text]
- Jenny, H. 1980. The soil resource: Origin and behavior. Springer-Verlag, New York.
- Lindsay, W.L. 1979. Chemical equilibria in soils. Wiley & Sons. New York.
- McBride, M.B. 1994. Environmental chemistry of soils. Oxford Univ. Press, New York.
- Meek, B.D., A.J. MacKenzie, and L.B. Grass. 1968. Effects of organic matter, flooding time, and temperature on the dissolution of iron and manganese from soil in situ. Soil Sci. Soc. Am. Proc. 32:634–638.
- Megonigal, J.P., S.P. Faulkner, and W.H. Patrick. 1996. The microbial activity season in southeastern soils. Soil Sci. Soc. Am. J. 60:1263–1266.[Abstract/Free Full Text]
- Owens, P.R. 2001. Inferring oxygen status in soils using iron metal rods. Ph.D diss. Texas A&M Univ., College Station, TX.
- Schwertmann, U., and R.M. Taylor. 1989. Iron oxides. p. 379–438. In J.B. Dixon and S.B. Weed (ed.) Minerals in soil environments. SSSA Book Series No. 1. SSSA, Madison, WI.
- Schwertmann, U., R.M. Taylor, and H. Stanjek. 1995. A lecture and demonstration for students on iron oxide formation, p. 11–15. In G.J. Churchman et al. (ed.) Clays controlling the environment. CSIRO Publishing, Adelaide, Aust.
- Soil Survey Staff. 1996. Soil survey laboratory methods manual. Soil survey investigations report 42, version 3.0. UDSA NRCS National Soil Survey Center, Lincoln, NE.
- Vepraskas, M.J., and S.P. Faulkner. 2001. Redox chemistry of hydric soils, p. 85–105. In J.L. Richardson and M.J. Vepraskas (ed.) Wetland soils. CRC Press, Boca Raton, FL.
TOP
ABSTRACT
INTRODUCTION
