SSSAJ Grow Your Career with SSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
Related Collections
Right arrow Laboratory Column Studies
Right arrow Mining Wastes
Right arrow Remediation
Right arrow Soil Chemistry
Soil Science Society of America Journal 67:778-786 (2003)
© 2003 Soil Science Society of America

DIVISION S-2—SOIL CHEMISTRY

Compost, Limestone, and Gypsum Effects on Calcium and Aluminum Transport in Acidic Minespoil

Frank J. von Willert and Richard C. Stehouwer*

Dep. of Crop and Soil Sciences, Pennsylvania State Univ., 116 ASI Building, University Park, PA 16802

* Corresponding author (rcs15{at}psu.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIAL AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Combining limestone (CaCO3) application with compost or gypsum (CaSO4·2H2O) can substantially increase subsoil Ca and decrease subsoil Al saturation in highly weathered acidic soils because of increased Ca mobility and formation of nontoxic Al-organic matter complexes. However, little is known about the effect of these surface amendments on subsoil chemistry in acidic minespoils that are high in SO4. This study used small segmented laboratory columns to analyze the effect of surface incorporated compost, CaCO3, and gypsum on Ca and Al chemistry in the highly acidic minespoil material (pH {cong} 2.5) below the zone of incorporation. Compost did not affect subsoil Al and Ca chemistry but caused a small increase in subsoil pH. Because of the high acidity in the spoil material CaCO3 solubility in the amended layer was high and was not increased by compost. Adding gypsum increased Ca leaching into the subsoil compared with CaCO3 alone, but extractable Ca was higher only after extensive leaching. Increases in extractable Ca in the subsoil were strongly correlated to decreases in extractable Al and Fe, indicating cation-exchange processes. However, significant losses of Al in unamended spoil columns and the small impact of CaCO3 and gypsum on total Al leaching indicated that Al chemistry was also influenced by a solid phase. Solubility calculations pointed to a jurbanitelike solid phase buffering Al activity in the subsoil, even after extensive leaching. Because of this buffering effect and the high acidity of the spoil material, none of the surface treatments would be expected to alleviate subsoil phytotoxicity in highly acidic minespoil material.

Abbreviations: AA, atomic absorption spectrophotometry • C, control • c, with compost • DOC, dissolved organic C • DOM, Dissolved organic matter • L, limestone treatment • LG, limestone and gypsum treatment • n, no compost • UAJA, University Area Joint Authority


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIAL AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
SULFIDE OXIDATION has transformed many former strip mine areas into acidic minespoils. The high acidity generated by the sulfide oxidation results in phytotoxic activities of monomeric Al (Al3+) throughout the spoil material. To establish a sustainable plant cover, successful reclamation should not only reduce acidity and Al3+ activity in the shallow amended surface layer, but also in the untreated subsoil material, which would increase the rooting depth. While a large volume of literature deals with the success of different methods for reclaiming a surface layer so that initial plant growth can be established (Sutton and Dick, 1987; Sopper, 1992), little attention has been given to the impact of these methods on the untreated spoil material below.

Surface treatment methods that beneficially affect subsoil acidity in naturally acidic soils are well studied and may point to possible methods to consider in acidic minespoils. It has been shown that surface incorporated CaCO3 does not significantly affect the untreated subsoil (Mathews and Joost, 1990; Pavan et al., 1984; Ritchey et al., 1980). Because the HCO-3 and OH- ions that are produced during the dissolution of CaCO3 quickly react with exchangeable H+ and Al3+ within the first few centimeters below the zone of CaCO3 incorporation, the comigrating Ca2+ ions cannot move further downwards and are quickly adsorbed (Sumner, 1995). Gypsum (CaSO4·2H2O) provides a more mobile source of Ca than CaCO3 because the SO2-4 ions can easily comigrate with Ca2+. Surface-applied or surface-incorporated gypsum can therefore significantly improve root-growing conditions by increasing exchangeable Ca and reducing exchangeable Al, without neutralizing any acidity (Oates and Caldwell, 1985; Pavan et al., 1984; Ritchey et al., 1980; Sumner et al., 1986; Sumner, 1995; Wendell and Ritchey, 1996). The effect can be very long lasting. Toma et al. (1999) found a significant and deep-reaching Ca increase in the subsoil of an Ultisol 16 yr after a one-time gypsum application. Surface application or surface incorporation of organic matter also decreased phytotoxic subsoil Al3+ activities because dissolved organic matter (DOM) that leached into the subsoil formed nontoxic Al–DOM complexes (Hargrove and Thomas, 1981; Hue 1992; Liu and Hue, 1996, Hue and Licudine, 1999). The combined application of CaCO3 and organic matter in lime-stabilized biosolids decreased subsoil acidity and increased subsoil Ca saturation, compared with CaCO3 alone (Brown et al., 1997; Tan et al., 1985; Tester, 1990). This effect was attributed to increases in Ca mobility caused by Ca–DOM complexes.

While exchangeable acidity and exchangeable Al3+ dominate the subsoil chemistry in most naturally acidic soils, acidic minespoils are characterized by high SO4 concentrations because of the oxidation of sulfides in the spoil material. Solid Al-OH-SO4 phases such as jurbanite [AlSO4(OH)·5H2O], alunite [KAl3(SO4)2(OH)6], or basaluminite [Al4(OH)10SO4·17H2O] might therefore impact Al3+ activities in untreated spoil material. Nordstrom (1982) showed that jurbanite is stable in natural waters up to a pH range of 3 to 5 and alunite up to a pH 4 to 7. Many other studies have shown the possibility of jurbanite controlling Al3+ activity in acidic soils (Evans, 1991; Menzies et al., 1994; Vogt et al., 2001) and in stream waters (Kram et al., 1995). Tin and Wilander (1995) showed that jurbanite precipitated in acid water leached from acid sulfate soils in Viet Nam. Based on solubility calculations, Monterroso et al. (1994) concluded that jurbanite controlled Al3+ activities in minespoil solutions. Additionally, it appeared that jurbanite controlled Al3+ activities in soils affected by acid mine waters (de Anta and Otero, 1994) and in acid mine pit lakes (Eary, 1999). Karathanasis et al. (1988) found that "the control of soluble Al by Al sulfate minerals was not affected by mineralogical and textural composition of soils and geologic strata" in watersheds affected by acid mine drainage. While they did not include jurbanite in their study, Prietzel and Hirsch (1998) note that the dissolution kinetics of Al-OH-SO4 in soils were slow, which might be the reason that Ludwig et al. (2001) found leachates from an open pit coal mine sediment undersaturated with respect to jurbanite and alunite. Karathanasis et al. (1988), who found very conclusive solution data evidence for the presence of a jurbanite solid phase, were unable to detect jurbanite peaks in x-ray diffraction experiments in the corresponding soil material. They attributed this to the presence of jurbanite in poorly crystalline or amorphous form. Possible Al-OH-SO4 phases in acidic minespoils will form during sulfide oxidation and may only exist in amorphous or poorly crystalline form, making mineralogical proof of their presence or absence very difficult.

While there are clear differences between naturally acidic soils and acidic minespoils, we hoped to observe similar beneficial impacts of CaCO3, gypsum, and compost on untreated subsoil spoil material as have been described for naturally acidic soils. In particular, Stehouwer et al. (1995) indicated that sewage sludge increased the mobility of Ca in minespoil material amended with gypsum containing flue gas desulfurization byproducts, but they did not examine the impact of this on untreated subsoil material. Because CaCO3 incorporated into the surface layer during minespoil reclamation will partially dissolve when neutralizing the acidity, it is questionable whether Ca solubility would be an issue with respect to beneficial effects of the surface on the subsoil.

This small-scale laboratory column study was conducted to assess the influence of surface incorporated gypsum, compost, and CaCO3 on subsoil chemistry in highly acidic minespoil material. The aim was to follow changes in soil solution and spoil material below the zone of incorporation of the surface amendments under an accelerated leaching regime, with an emphasis on Al and Ca, while focusing purely on chemical processes, eliminating any influence of plant and root growth.


   MATERIAL AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Spoil and Compost Material
We collected acidic minespoil material from an abandoned mine land site in Jefferson County, PA and screened the mainly shale material to <20 mm. This fraction contained 60% rock fragments (2–20 mm). We carefully crushed an air-dried subsample in a mortar until all of it passed through a 2-mm sieve. The resulting material had 78.5% sand, 9.7% silt, and 11.8% clay. A Mehlich-3 extract of the ground material contained 0.8 mmolc kg-1 K+, 10.8 mmolc kg-1 Mg2+, and 13.6 mmolc kg-1 Ca2+. The pH in 1 M KCl extracts was 2.5. Exchangeable acidity and exchangeable Al3+, determined by extraction with 1 M KCl (1:5 solid/solution suspension with a 1-h shaking followed by filtration) averaged 77 ± 4 and 43 ± 3 mmolc kg-1, respectively. The SMP buffer pH method (Eckert and Sims, 1995) estimated a total acidity of 285 mmolc kg-1.

For the compost amendments we used a biosolids compost from the University Area Joint Authority (UAJA) composting facility in State College, PA. This compost is generated by composting undigested, dewatered sludge from the primary and secondary clarifiers of the UAJA wastewater treatment facility with hardwood sawdust. The composting process is carefully controlled and takes about 26 d. Lime is not added at any point of the process. The compost we used for our experiments was ground to <2 mm. It had a pH of 6.8 and contained 430 g kg-1 C, 22 g kg-1 N, 1.26 g kg-1 Ca (63 mmolc kg-1 Ca2+). The ash content was 110 g kg-1.

Leaching Experiments
We used standard 60-mL syringes to construct small, segmented columns that could be leached at a constant flow rate using a mechanical vacuum extractor (Centurion Model 24, Centurion International, Lincoln, NE). We employed this segmented column approach so that we could easily sample soil solution immediately below the amended surface material and at two different depths in the untreated subsoil material.

First Experiment–Incorporated Compost
Each soil column consisted of three individual syringe segments ( Fig. 1) . The first column segment was filled with 48 g of spoil material, combined with one of three Ca sources (Control = C, 13 g kg-1 CaCO3 = L, 13 g kg-1 CaCO3 + 22.3 g kg-1 gypsum = LG); each with (c) and without (n) 68.8 g kg-1 compost, resulting in six different treatments (Cc, Cn, Lc, Ln, LGc, LGn). We will refer to this first segment with the amended minespoil material as the surface soil. The application rates of the amendments were the same as those used in a large column experiment with the same material, and corresponded to three times the exchangeable acidity in the surface material (von Willert, 2002). Each of the other two segments per column was filled with 32 g of untreated spoil material. We will refer to these segments as the subsoil. All segments were brought up to field capacity using deionized water and were allowed to equilibrate overnight. The mechanical vacuum extractor was used to leach the columns by alternating a 40-mL leaching and a 5-mL sampling scheme. For the 40-mL leaching scheme we first leached 40 mL of deionized water through the surface soil segment. The leachate from this segment was then leached through the first subsoil segment, and the leachate generated there was leached through the second subsoil segment. The 5-mL sampling scheme allowed us to follow solution chemistry changes at different depths and involved leaching three 5-mL batches of deionized water through (i) the surface soil segment only, (ii) the surface soil segment and the generated leachate through the first subsoil segment, and (iii) sequentially through all three segments (Fig. 1). The leaching rate was 20 mL h-1 (38 mm h-1) and a minimum of 3 h passed between consecutive leachings of the same segment. The experiment was performed in duplicate with one replicate at a time because of limited space on the vacuum extractor.



View larger version (37K):
[in this window]
[in a new window]
  		Fig. 1. Schematic diagram of one segmented soil column. Leachate was moved through the soil using a mechanical vacuum extractor at a flow rate of 38 mm h-1. Leaching was conducted by alternating six times a) leaching of 40 mL of solution sequentially through all three segments and b) leaching of three 5-mL batches (i) through the surface soil segment only, (ii) through the surface soil segment and then through the first subsoil segment, and (iii) through all three segments. In the second experiment nine consecutive batches of 40 mL were sequentially leached through all three segments before compost was added and the 40-mL leaching and 5-mL sampling scheme was performed.

 
All collected solutions were filtered through 0.45-µm membrane filters (Durapore® PVDF, Millipore, Bedford, MA) and stored at 4°C until further analysis. Aluminum and Ca concentrations were measured by atomic absorption spectrophotometry (AA), dissolved organic C (DOC) as nonpurgeable organic C on a TOC-5000A analyzer (Shimadzu Scientific Instruments, Columbia, MD), and pH using a Ross Semi-Micro pH electrode connected to a 520A pH meter (Orion Research, Beverly, MA).

At the end of the experiment the soil in the syringes was extracted with 1 M KCl (1:10 solid/solution ratio with a 1-h shaking followed by filtration). Aluminum, Ca, and Fe concentrations were measured by AA, pH was measured as above.

Second Experiment–Preleaching and Surface Applied Compost
In a second experiment the columns were preleached before we surface-applied compost and collected solution samples. The soil columns were prepared as described for the first experiment, but only three treatments were examined: control (C), CaCO3 (L), and CaCO3 + gypsum (LG). After adding CaCO3 and gypsum to the surface soil segments at the same ratios as in the first experiment, the columns were brought to field capacity with deionized water and equilibrated overnight. Then the three segments of each column were sequentially leached with nine 40-mL doses of deionized water, equivalent to the total leachate volume in the first experiment. As before, a minimum of 3 h passed between consecutive leachings of the same segment. No samples were taken during the preleaching phase, assuming that the data of the compost-free treatments from the first experiment were a good proxy for the conditions during the preleaching period. After the preleaching, a round of 5-mL samples was collected as described above to measure the solution chemistry before compost addition. Then we applied 3.3 g of compost, corresponding to a rate of 68.8 g kg-1, to the top of all surface segments, not mixing it with the preleached soil. The compost was moistened to field capacity with deionized water and equilibrated overnight. Then the same alternating 40-mL leaching and 5-mL sampling scheme as described above was repeated six times. Consequently, the total leachate volume in the second experiment was twice as high as in the first experiment. This second experiment was performed with three simultaneous replicates. At the end the soil in each segment was extracted with 1 M KCl. Leachates and extracts were analyzed as described for the first experiment.

Statistical Analyses
All statistical analyses of the measured response variables in the two subsoil columns were conducted with S-Plus (Mathsoft, 1999). The statistical layout for the first experiment was a randomized block factorial design with repeated measures on the segment factor and complete randomization for the Ca source and compost factors. Because of lack of degrees of freedom no interaction was assumed between the other factors and the blocking factor (replication). The second experiment was analyzed as a repeated measures factorial design with repeated measures on segment and with complete randomization of Ca source.

Analysis of variance followed the general procedures outlined in Neter et al. (1996) and the recommendations for repeated measures models in Girden (1992) and Kirk (1995). The ANOVA tables for these repeated measures designs contained two error strata, one for the between columns effects (Ca source and compost) and one for the error associated with the repeated measures factor (segment), the within columns effect. Main factor level differences were tested with the appropriate error stratum mean square error. If there was interaction between the repeated measures variable (segment) and the other main factors, separate ANOVA tables were calculated for each segment to obtain the appropriate error variances for testing factor level effects.

In the first experiment seven simultaneous tests had to be conducted to test three main factor level effects, three two-way interactions, and one three-way interaction. At a family confidence level of {alpha} <= 0.1 each of these seven tests was significant at {alpha}i <= 0.015 according to the Kimball inequality (Neter et al., 1996). Mean differences were tested at a family confidence level of {alpha} <= 0.1, with individual test levels calculated using the Bonferroni inequality (Neter et al., 1996). In the second experiment, ANOVA involved three tests for two main factor effects and one interaction. To be considered significant at a family confidence level of {alpha} <= 0.05 each individual test had to be significant at {alpha}i <= 0.017 following the Kimball inequality. Mean differences were tested at a family level of {alpha} <= 0.05, using the Bonferroni method to calculate individual test levels.

Stability Diagrams
To evaluate whether the sampled leachate solutions were influenced by Al-OH or Al-OH-SO4 solid phases, we needed to calculate Al3+ and SO2-4 activities. We did not directly measure SO4 concentrations, but assumed that because of the nature of acidic minespoils SO2-4 was the dominating anion in all solutions. Total SO4 concentrations could then be estimated as [SO4]total = 0.5(3[Al]total + 2[Ca]total), where brackets indicate total concentrations in mol L-1. We then used the chemical speciation program ECOSAT to calculate Al3+ and SO2-4 activities in all solution samples, using the measured total Al and Ca concentrations, the measured pH, and the calculated total SO4 concentrations as input parameters. The speciation model accounted for the formation of AlSO+4 and CaSO04 complexes.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Compost Effects
Incorporating only compost increased pH in the surface soil at the end of the first experiment from 2.81 to 3.56 (Table 1), because the whole compost could buffer acidity. In the second experiment the soluble organic matter that leached from the surface applied compost into the surface soil segment did not provide sufficient buffering capacity and the final soil pH of the compost only treatment was not different from the compost free control treatment in the first experiment (Tables 1 and 2).


View this table:
[in this window]
[in a new window]
  		Table 1. Extractable cations (1 M KCl) and cumulative mass balance for the first experiment. Means of two replicates. ANOVA results shown in Table 3.

 

View this table:
[in this window]
[in a new window]
  		Table 3. ANOVA results for the data in Table 1 (three factors with repeated measures on segment). Only data for the subsoil segments were included in the analysis, with two replicates per treatment.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Extractable cations (1 M KCl) and sampling period cumulative mass balance for the sampling period of the second experiment. Means of three replicates. ANOVA results shown in Table 4.

 

View this table:
[in this window]
[in a new window]
  		Table 4. ANOVA results for the data in Table 2 (two factors with repeated measures on segment). Only data for the subsoil segments were included in the analysis, with three replicates per treatment.

 
The only conclusive effect of compost on the subsoil besides a slight increase in pH was a reduction of extractable Fe (Tables 1 and 3), which can be attributed to the formation of Fe–organic matter complexes that can be very strong (Schnitzer and Skinner, 1964). A beneficial effect of DOM could have been an increase in Ca solubility and thus increased transport of Ca into the subsoil (Stehouwer et al., 1995, Hue and Licudine, 1999). However, even in treatments that only received CaCO3 in the first segment, Ca solubility in the surface soil was so high that DOM did not have any effect on subsoil Ca chemistry (Table 2, Fig. 2) .



View larger version (42K):
[in this window]
[in a new window]
  		Fig. 2. Leachate Al (left panel) and Ca (right panel) concentrations. Data for both experiments are printed in one graph because the compost free treatments in first experiment (open symbols) can be assumed to represent the leachate conditions during the preleaching phase (360 mL) in the second experiment, during which no samples were taken. The first data point in the second experiment was taken after the preleaching, but before the surface application of compost, which is indicated by the vertical arrows. Means of two replicates for first experiment and three replicates for the second experiment.

 
Our own measurements with DOM derived from the compost used in this experiment at pH 3 indicated an Al3+ complexation capacity of 0.4 molc Kg-1 C (von Willert, 2002). Combining this with the leachate DOC concentrations, we estimated that in a few samples a maximum of 25% of free Al3+ was complexed by DOM in the subsoil solutions of compost containing treatments. In most leachate samples the Al binding capacity of the DOM was well below 10% of the free Al3+ concentration. Based on these findings, we conclude that formation of Al-DOM complexes in the subsoil solutions did not significantly affect Al3+ activity.

Calcium Source Effects
pH
The amount of CaCO3 added was sufficient to ensure an above neutral pH in the surface soil at the end of both leaching experiments (Tables 1 and 2). Adding CaCO3 to the surface soil increased solution pH (data not shown) and soil pH in the subsoil in both experiments, while gypsum had no additional effect (Tables 3 and 4). The effect on subsoil pH decreased with increasing depth. Despite being significant, the maximum differences between L and LG treatments and the control were <0.5 pH units, too small to alleviate Al3+ toxicity.

Calcium
The differences between total Ca2+ added (260 mmolc kg-1 for the L treatments, and 520 mmolc kg-1 for the LG treatments) and extractable Ca2+ in the surface soil at the end of the first experiment was very close to the Ca2+ losses calculated from the cumulative mass balance (Table 1). We therefore assumed that the cumulative mass balance calculations were reasonable approximations of the actual mass balances. Based on this assumption our data indicate that about half of the added Ca leached out of the surface soil in the L and LG treatments during the first experiment (Table 1).

In the L treatments, the Ca concentration in the leachate from the surface soil dropped from an initial value of more than 30 mmolc L-1 to about 10 mmolc L-1 in the first experiment and further down to below 5 mmolc L-1 in the second experiment (Fig. 2). High Ca solubility in these treatments was caused by the acid-promoted dissolution of the added CaCO3. In the subsoil of the L treatments Ca concentrations continually decreased as well. They also decreased with depth during the first experiment, showing that Ca was sorbed in the subsoil. In the second experiment leachate Ca continued to decrease, but concentrations did not change with depth (Fig. 2), indicating that no more Ca sorption was taking place. In the LG treatments, Ca concentrations in surface soil leachates stayed at values above 25 mmolc L-1, the solubility of CaSO4·2H2O, for the duration of the first experiment and only decreased during the second experiment. In the subsoil, leachate Ca concentrations remained fairly constant during the first experiment but decreased with depth (Fig. 2), indicating that Ca was sorbed. While Ca concentrations in the subsoil leachates decreased during the sampling phase of the second experiment, decreasing leachate concentrations with increasing depth showed that Ca sorption was ongoing in the subsoil of the LG treatments.

Cumulative mass balances calculated from the leachate concentrations showed that more Ca was sorbed in the subsoil of LG treatments than in the subsoil of L treatments in both experiments (Tables 3 and 4). However, in the first experiment there was no difference in extractable subsoil Ca between L and LG treatments (Table 3). In the second experiment, where cumulative mass balance indicated that after preleaching no more Ca sorbed in the subsoil of the L treatment while sorption was ongoing in the LG treatment, extractable subsoil Ca was higher in the LG than in the L treatment (Tables 2 and 4). The mass balances also show that more than half of the Ca passed through the subsoil without being sorbed (Table 5).


View this table:
[in this window]
[in a new window]
  		Table 5. Comparison of the amount of Ca leached from the surface soil and the amount sorbed in the subsoil. Data are based on leachate measurements and were calculated from the cumulative mass balance data in Tables 1 and 2.

 
Overall, there is evidence that some exchange of Al for Ca took place in the subsoil of all treatments but the control. Besides H+, Ca2+ sorbed in the subsoil would most likely have replaced Al3+ and Fe3+. Figure 3 shows that a very high correlation existed between extractable Ca2+ charges and the sum of extractable Al3+ and Fe3+ charges for all treatments except the control. The slopes of the structural relations indicate that increases in Ca2+ were larger than the reduction of Al3+ and Fe3+. The reason for this may be that some of the sorbed Ca2+ replaced extractable H+. On the other hand, 1 M KCl extractable Al may include some contribution from dissolving solid phases in addition to exchangeable Al. Some of the Ca retained in the subsoil might also have been in the form of a Ca-Al-OH-SO4 solid phase, about which is very little known (Myneni et al., 1998). However, the strong correlations in Fig. 3 are a good indicator that Ca that leached from the surface soil into the subsoil replaced extractable Al and Fe.



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 3. Correlation between the sum of extractable Al and Fe and extractable Ca in the subsoil. Lines calculated by structural analysis assuming the error for both variables was approximately equal (Webster, 1997). C = Control, L = CaCO3, LG = CaCO3 and gypsum.

 
Aluminum
During the first experiment, leachate Al concentrations decreased rapidly in all subsoil segments within the first 100 mL of leachate volume, but never dropped below detectable levels (Fig. 2). More Al leached from the LG treatments than from the control treatments (Table 3). Aluminum concentrations during the sampling phase in the second experiment were similar in all treatments and decreased with increasing leachate volume (Fig. 2) and neither CaCO3, nor gypsum, nor compost in the surface soil affected total Al leaching in the subsoil (Table 4).

Adding a Ca source to the surface decreased extractable subsoil Al compared with the control in both experiments. There was a deeper reaching effect when gypsum was added to the surface (Tables 3 and 4). Mass balances calculated from leachate Al concentrations, on the other hand, provided evidence that subsoil Al was not only affected by exchange for Ca. While the control treatment lost significant amounts of Al during the sampling phase of both experiments, extractable Al at the end of the second experiment had not changed compared with the first (Tables 1 and 2). Moreover, increasing amounts of Al were lost from the control treatment with increasing depth, but the extractable Al3+ remained the same at the end of the experiment (Tables 1 and 2). Clearly, mineral dissolution influenced Al leachate concentrations in the control treatments. Consequently it may have been an important process in the subsoil of L and LG treatments as well. Since mineral dissolution should add to any Al exchanged by Ca, one might expect higher subsoil leachate Al concentrations in the L and LG treatments than in the control. However, Al concentrations differed only slightly between control and L and LG treatments (Fig. 2). There was no significant effect of Ca source on subsoil Al loss during the sampling phase of the second experiment, despite continuing Ca sorption in the LG treatment (Table 4). This agrees with observations by Oates and Caldwell (1985) and Sumner et al. (1986) who found significant reduction in exchangeable Al in naturally acidic subsoils after surface application of gypsum without finding an accompanying increase in leachate Al, which they attributed to polymerization of Al in the form of Al-OH or Al-OH-SO4 solid phases.

Figure 4 shows that in both experiments subsoil leachates of all treatments and surface soil leachates of the control treatments were undersaturated with respect to diaspore (AlOOH), one of the least soluble Al-hydroxide solid phases (Lindsay and Walthall, 1996). Leachate Al3+ activity in most samples was most closely approximated by jurbanite solubility (Fig. 4). The effect of compost seen in some samples of the Cc, Lc, and LGc treatments (Fig. 4a and 4b) might have been related to the additional alkalinity provided by the soluble organic matter and the possibility that the soil solutions were not in full equilibrium with the solid phase. As we pointed out above, formation of Al-DOM complexes most likely did not play a very significant role. In the lower subsoil segments of both experiments, where DOC concentrations were lowest and Al concentrations were highest, solution data were very closely distributed around the line representing jurbanite solubility (data not shown).



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4. Solubility diagrams for (a) first segment, first experiment, (b) second segment, first experiment, (c) first segment, second experiment, (d) second segment, second experiment. Parentheses indicate activities. Al3+ activities calculated from total Al concentrations using ECOSAT, accounting for AlSO+4 and CaSO04 complexes. Error bars are for log(SO2-4) - 1 and log(SO2-4) + 1 respectively. C = Control, L = CaCO3, LG = CaCO3 + gypsum, n = no compost, c = 68.8 g kg-1 compost.

 
Because we only measured pH, Al3+, and Ca2+ concentrations in the leachates and we had to approximate SO2-4 concentrations, and because solubility calculations assume equilibrium, which might not have existed because of the fairly rapid leaching, the data in Fig. 4 have to be interpreted carefully. However, as we pointed out in the introduction, there is ample evidence in the literature for jurbanite control of Al3+ activity under acidic and high SO4 conditions. Despite the approximate nature of the solubility calculations, these other reports support the assumption that subsoil Al3+ activity was not only affected by exchange reactions, but also by dissolution and precipitation of a jurbanitelike solid phase. This constitutes a great difference to highly weathered acidic soils, where most of the buffering of Al3+ activity is influenced by the Al3+ saturation of the exchange sites. It might therefore explain, why subsoil conditions improved only up to a certain point and why, even with high Ca solution concentrations, significant amounts of extractable Al remained in the subsoil of the LG treatments (Tables 1 and 2). If a jurbanitelike solid phase buffered Al3+ solution activity in the subsoil, the amount of extractable Al would not only be controlled by the transport of Ca from the surface soil into the subsoil. A lasting improvement of subsoil Al toxicity could then only be expected after all of the hypothesized jurbanitelike solid phase dissolved or if sufficient alkalinity is transported downward to substantially raise the subsoil pH. The current experiment does not allow us to estimate how long complete dissolution of the jurbanitelike solid phase would take, but the small changes we observed after extensive leaching indicate that none of the tested surface treatments may be powerful enough to reduce subsoil Al3+ toxicity.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
In highly acidic minespoil, extractable Al in the subsoil decreased only to a certain extent by exchange with Ca. Subsoil solution Al3+ activity was also influenced by a jurbanitelike solid phase and consequently considerable amounts of extractable Al remained in the subsoil, even after extensive leaching with soil solution high in Ca. Limestone solubility in the amended surface soil was high due to acidity promoted dissolution, which is why compost did not increase Ca mobility. Adding gypsum in addition to CaCO3 did not affect subsoil pH, but ensured more Ca sorption in the subsoil and a deeper reaching effect, because Ca solution concentrations stayed higher for a longer time. Control of subsoil Al3+ solution activity by a jurbanitelike solid phase means that this solid phase needs to be dissolved before Ca-exchange reactions can significantly reduce subsoil Al3+ toxicity. However, it can hardly be desirable in the field to increase leaching rates to achieve this, as that might increase problems with acid mine drainage. Therefore the reclamation effort in highly acidic minespoil material must ensure the transport of alkalinity into the subsoil. Probably the only way to significantly reduce Al3+ activity and restore conditions favorable to root growth is the deep incorporation of a liming agent. Incorporation of organic matter in the surface may aid insofar as released DOM provides additional alkalinity to the subsoil, but as long as a hypothesized jurbanitelike solid phase buffers Al3+ activities in the subsoil, it cannot aid in detoxifying subsoil solution Al3+.


   ACKNOWLEDGMENTS
 
This research was supported by Research Grant Award No. IS-2758-96R from BARD–The United States–Israel Binational Agricultural Research and Development Fund.

Received for publication April 14, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
Right arrow Search for Related Content
PubMed
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by von Willert, F. J.
Right arrow Articles by Stehouwer, R. C.
Related Collections
Right arrow Laboratory Column Studies
Right arrow Mining Wastes
Right arrow Remediation
Right arrow Soil Chemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome