| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a 27 Drew Ave., Howick 3290, South Africa
b KwaZulu-Natal Dep. of Agric., Cedara College, Private Bag X9059, Pietermaritzburg 3200, South Africa
farina{at}nitrosoft.co.za
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
0.4 units, by similar increases in 
pH (pHw - pHs), by depressions in exchangeable acidity of as much as 1.5 cmolc L-1, and by decreases in acid saturation of more than 30%. The rate of subsoil amelioration was, however, much slower than that reported in more intensely weathered soils of similar texture. Only in the sixth season were benefits evident in the 0.60- to 0.75-m horizon, and acidity in the 0.75- to 0.90-m horizon actually increased significantly. It is speculated that this resulted from NO3 accumulation and ionic strength–induced dissolution of interlayer Al. These findings indicate that acid-subsoil amelioration in soils with Al-hydroxy–interlayer minerals requires greater quantities of gypsum than soils that are dominantly kaolinitic. |
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Choice of the best approach to adopt is necessarily governed by economic considerations and is also strongly soil dependent. For example, some soils, particularly those that are sandy or have been anthropogenically acidified, may not be responsive to gypsum (Black and Cameron, 1984; Horsnell, 1985; Keerthisinghe et al., 1991), while deep tillage is undesirable on soils with dense subsoils (Tupper et al., 1987; Coventry, 1991). Also, there is evidence to suggest that the gypsum application rates used with such success in the tropical savannas of Brazil (Ritchey et al., 1995) have few, if any, beneficial effects on the less intensely weathered, but equally acidic Oxisols and Ultisols of South Africa (Farina, 1997). Similarly, the quantities of lime required to elevate topsoil pH levels sufficiently to promote downward movement of alkalinity in intensely weathered soils of the humid and subhumid tropics (Sanchez, 1977; Lathwell, 1979; McKenzie et al., 1988) are much less than those needed on soils of similar acidity in less weathered environments (Farina and Sumner, 1979; Farina, 1997). In the latter environments, soils equally acidic to their textural and morphological counterparts in the moist tropics usually contain greater absolute amounts of exchangeable Al, as well as substantial quantities of potentially active Al associated with mixed-layer clay mineralogy. This reserve acidity constitutes an important component of lime buffer capacity in the pHw range 4 to 6 (Juo and Kamprath, 1979) and it is conceivable that replenishment of Al3+ from initially nonexchangeable sources is also responsible for the very high gypsum requirements recorded on some such soils (Farina, 1997). Marked variations in the efficacy of lime and gypsum as ameliorants, even on soils of similar clay content, bedevils the management of subsoil acidity in environments outside the humid tropics, a situation further aggravated by a paucity of sufficiently descriptive reports on long-term field studies.
In a companion paper (Farina et al., 2000), we discussed the effects of heavy applications of lime (both topsoil-incorporated and deep-placed) and gypsum on maize (Zea mays L.) growth for 10 seasons. A comparison of the various treatments imposed indicated that the basal application of lime (15 Mg ha-1) plus 10 Mg ha-1 of gypsum was considerably more effective than the basal lime application either conventionally incorporated or incorporated to a depth of 0.50 m, more effective than the large extra applications of lime incorporated in vertical bands to depths of about 0.80 m, and more effective even than the basal lime application plus a further 10 Mg ha-1 incorporated to 0.50 m. The objective of this paper is to report on the long-term chemical changes in the soil profile that were responsible for differences in growth. Hopefully, this discussion will help to expand the pool of knowledge needed for the optimal management of subsoil acidity in the diverse environments in which it has been identified.
|
Materials and methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
To obtain information regarding the dynamics and longevity of the gypsum treatment, conventionally limed and gypsum-treated plots were depth sampled after harvest each season. In the 1982–83 and 1983–84 seasons, samples were obtained to 0.60 and 0.75 m, respectively. Thereafter, the profile was sampled to a depth of 0.90 m.
In the interests of brevity, comparisons were made using differentials (value of the chemical parameter in the gypsum treatment minus the value in the conventional treatment). Since long-term trends were of primary concern and a very considerable number of individual comparisons was possible, measures of statistical significance are not reported. However, such measures between treatments, both within and across seasons, were obtained by pooling data from the last eight seasons (constant sampling depth) and analyzing them as a treatments (2) x years (8) x depth (6) factorial design using Genstat V Release 2.2 (Lawes Agricultural Trust, 1983). In the discussion to follow, the term significant has been exclusively used to indicate, where necessary, between or within season comparisons that differed at a probability level of 
0.05.
|
Results and discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
|
0.50 m), the possibility of some movement cannot be excluded, but such movement must necessarily have been very limited. In view of the fact that the highest topsoil salt pH attained with this treatment was only 4.68 (Farina et al., 2000), failure to detect meaningful movement of alkalinity is perhaps not surprising (Helyar, 1991). The inability of such large applications of lime to ameliorate subsoil acidity rather dramatically demonstrates the futility of attempting to use surface-incorporated lime as a subsoil ameliorant on soils of similar clay content and mineralogy to that studied here. There are patently large differences in this respect between such soils and their textural counterparts in more intensely weathered environments, where more modest lime applications have been shown to have substantial effects on deep-seated acidity (Sanchez, 1977; Lathwell, 1979; McKenzie et al., 1988). Comparison of these data (Table 1) with the chemical properties of the virgin soil (Farina and Channon, 1988a) indicate that considerable vertical movement of Ca and Mg occurred in all the lime-only treatments, but no significant differences were evident between treatments below the original depth of lime incorporation. More importantly, likely due to increases in electrolyte concentration (Farina, 1997), marked simultaneous increases occurred in the levels of exchangeable acidity in the deepest subsoil horizons. Consequently, the net effect on acid saturation was negligible. This illustrates quite well the dangers associated with the assessment of lime movement in terms of cation accumulation in deeper horizons. Such movement may well have agronomic significance where subsoils are Ca-deficient (Ritchey et al., 1980), but is not necessarily relevant in situations where root penetration is restricted by acidity.
Temporal Effects of Gypsum
The progressive effects of gypsum on profile chemical properties for this study are illustrated in Fig. 1 and 2
. Clearly, substantial changes in profile properties occurred after the 4-yr period reported on previously (Farina and Channon, 1988b) and useful information is provided with regard to the rate of gypsum movement, longevity of its ameliorative action, and its probable mode of action in this soil.
|
|
In the two uppermost soil horizons, apparently anomalous increases in the Ca status of gypsum-treated plots occurred during the 1986–87 and 1991–92 seasons. In both cases, these increases were statistically significant. It seems probable that the effect noted in 1986–87 resulted from exceptionally high rainfall during 1985–86 (Farina et al., 2000) and a resultant surge in the dissolution and downward movement of gypsum. Subsequent increase in the differential between gypsum-treated and conventionally limed plots at the end of the 1986–87 season conceivably reflected the effects of further, more modest plow-layer dissolution accompanied by a lower removal rate to subsoil horizons. Rainfall cannot easily be invoked as a causal agent for the increased differentials noted in 1991–92 and it is considered more likely that the effect was due to reduced sorption of gypsum in previously gypsum-treated plots following the high application of single superphosphate in 1991 (Farina et al., 2000).
The effects of gypsum on subsoil Mg accumulation, while understandably much less in absolute terms, were similar to those on Ca. Little movement beyond 0.75 m occurred and redistribution after the eighth season was small (Fig. 1). The negative effects of gypsum on the plow-layer Mg status were very marked; but it is noteworthy that the topsoil Mg status was not reduced to a level that would be considered critical (Table 1) and that Mg did not move beyond the zone of improved rooting (Farina et al., 2000). This suggests that on soils with effective cation-exchange capacities similar to that studied here (5 cmolc L-1), the often cautioned-against negative effects of gypsum on Mg availability might not be meaningful. This observation is supported by findings obtained during a 7-yr period in an adjacent trial involving differential annual applications of Mg (0, 50, and 100 kg ha-1), once-off applications of gypsum (0, 5, and 10 Mg ha-1), and where Ca(OH)2 rather than Mg-containing lime was used to maintain satisfactory levels of topsoil acidity (M.P.W. Farina, unpublished data, 1997). There have been significant maize-yield and plant-compositional responses to Mg, but there has been no evidence of a Mg x gypsum interaction.
Year-on-year exchangeable soil K data are not shown, but as is evident in Table 1, gypsum had no effect on K movement. The significant reduction in the topsoil (0–0.15 m) K status of gypsum-treated plots almost certainly resulted from improved growth in these plots and increased K removal.
Extractable and Exchangeable SO4–Sulfur
The progressive downward movement of SO4–S to subsoil horizons is clearly evident in terms of both extractable and exchangeable SO4 (Fig. 1). Equally apparent is the fact that appreciable redistribution of SO4 continued throughout the study and that conversion of SO4, via either sorption or precipitation reactions (Rajan, 1978; Hue et al., 1985; Shainberg et al., 1989), to a form not exchangeable with dilute CaCl2 has been particularly marked in subsoil horizons.
A striking feature of these data, when comparisons are made with Brazilian gypsum studies on soils of similar clay content, is the slow rate of SO4 movement into the subsoil in this study. Even though Ritchey et al. (1995) reported appreciable SO4 movement to depths greater than 1 m within 2 yr of applying 6 Mg ha-1 of CaSO4·1/2H2O, in this study SO4 accumulation below 0.75 m was nonsignificant after 10 seasons; even in the 0.45- to 0.60-m horizon, SO4 did not peak until the seventh season. Moreover, in the Brazilian study there was little evidence of residual SO4 in the topsoil (0–0.15 m) after only two seasons, while in this study the differentials in topsoil-extractable SO4 levels remained statistically significant for 5 yr. This suggests that, in part at least, differences in precipitation and the rate of water throughflow played a role. Soils in the central savanna of Brazil are typically deep and well-drained and seasonal precipitation is about twice that recorded in this study (Goedert, 1983).
It also seems probable, however, that clay mineralogical differences between the soil studied here and those more typical of the moist tropics played a role. Mixed-layer clays constitute about 30% of the clay faction in this soil, but may be absent in intensely weathered Brazilian soils (J.C. Hughes, personal communication, 1997). In this soil, the mixed-layer clay component is associated with considerable quantities of nonexchangeable Al (in horizons below 0.30 m, up to 11 cmolc Al L-1 are released after four extractions with 0.5 M CuCl2) and it is reasonable to assume that this reserve acidity might replenish Al3+ in the soil solution (Juo and Kamprath, 1979) and so buffer the self-liming effect of gypsum. Also, gypsum-induced increases in pH (Table 1 and discussion to follow) would be expected to exert a greater effect on cation-exchange capacity in soils of mixed clay mineralogy than in soils that are dominantly kaolinitic (de Villiers and Jackson, 1967). Appreciable differences in pH (pHw - pHs) between 0.30 and 0.75 m (the zone of maximum SO4 accumulation [Fig. 1 and 2]) suggest that some retardation of gypsum movement could have resulted in this fashion.
Findings obtained in an adjacent trial, where variable rates of gypsum up to 16 Mg ha-1 have been used, suggest that the former possibility (release of Al from nonexchangeable sources) may have played a dominant role. In that trial, the 0.75- to 0.90-m horizon acidified markedly over 6 yr (Farina, 1997). This increase in acidity was not associated with the downward movement of surface-generated acidity but was speculated to have resulted from Ca(NO3)2 accumulation, electrolyte-induced hydrolysis of Al and Fe on edge sites (Chang and Thomas, 1963; Thomas and Hargrove, 1984), and from dissolution of interlayer Al (Weaver, 1972; Corti et al., 1997). Similar increases in extractable NO3 were evident in this study (Table 1) and when comparisons are made with analyses conducted on the soil in its virgin state (Farina and Channon, 1988a) and during 1984–85 to 1991–92 (Table 2) , there is clear evidence that here, too, acidity levels in the deepest horizon of both conventional and gypsum-treated plots increased with time.
|
With respect to the above somewhat speculative discussion, it is particularly interesting to note that no evidence of SO4 consumption against Ca(H2PO4)2 extraction existed in the study reported on by Ritchey et al. (1995) in Brazil. Also, unpublished data from the variable-rate gypsum study discussed by Farina (1997) indicate that the introduction of soybean [Glycine max (L.) Merr.] in the place of maize for three seasons (1994–1996) and suspension of inorganic N applications resulted in large reductions in subsoil NO3 and acidity levels, appreciable increases in pHw, and very marked increases in extractable SO4. In the 0.60- to 0.75-m horizon, where these effects were most evident, exchangeable NO3 and Al + H levels decreased from 0.59 to 0.14 and 4.16 to 3.02 cmolc L-1, respectively; pHw increased from 4.43 to 4.73; and extractable SO4 increased from 0.28 to 1.35 cmolc L-1. Total extractable SO4 to a depth of 0.90 m in the profile increased by 35% and for the first time since the experiment was initiated in 1985, statistically significant benefits from gypsum were measurable in the 0.75- to 0.90-m horizon. At the highest rate of lime and gypsum application (Farina, 1997), a 50% decrease in the level of extractable NO3 (from 0.42 to 0.21 cmolc L-1) was accompanied by increases in pHw (from 4.43 to 4.65) and extractable SO4 (from 0.01 to 0.73 cmolc L-1), and decreases in exchangeable acidity (from 4.95 to 4.47 cmolc L-1) and acid saturation (from 78 to 64%). Such effects are entirely compatible with proposals made by Adams and Rawajfih (1977) and suggest that on acid soils containing appreciable quantities of interlayer Al, for as long as the pH remains sufficiently low, Al dissolution will remove SO4 from the soil solution via precipitation of hydroxy-sulphate minerals with little or no net ameliorative effect. This, in turn, suggests that the efficacy of gypsum as a subsoil ameliorant will depend on management practices that influence salt movement and accumulation in subsoil horizons. Nitrogen applications are of singular importance in this regard and in view of the costs associated with gypsum use, this is an aspect which warrants further investigation.
Exchangeable Acidity, Percent Acid Saturation, and pH
Exchangeable acidity differentials between gypsum-treated and control plots (Fig. 1) indicate that gypsum had a marked beneficial effect on the rooting environment. Here too, the time dependency of gypsum's ameliorative effect is clearly evident, the mean differential between treatments having increased from 0.41 cmolc L-1 in 1984–85 to 0.85 cmolc L-1 in 1991–92. Clearly, except for the deepest horizon, there is a reasonably good relationship between the depression in the level of exchangeable acidity and the quantity of SO4 sorbed against exchange with CaCl2. It should be noted, however, that the picture depicted in terms of treatment differences fails to reflect the negative effects of high N use already discussed. In the 0.75- to 0.90-m horizon, acidity levels actually increased significantly with time under both treatments (Table 2), and the favorable differential depicted in Fig. 1 in reality reflects slower acidification in gypsum-treated plots rather than an improvement in the soil condition per se. Similarly, in the 0.45- to 0.60-m and 0.60- to 0.75-m horizons, where gypsum significantly depressed exchangeable acidity levels during the period 1984–85 to 1991–92 by 0.92 and 0.55 cmolc L-1, respectively (Tables 1 and 2), the apparent benefit (Fig. 1) was exaggerated by the fact that in control plots, acidity levels had simultaneously increased. Thus, while the exchangeable acidity differentials depicted in Fig. 1 reflect the benefits of gypsum under the experimental conditions that existed in this study, they might have differed appreciably had N inputs more nearly matched crop requirement. It is noteworthy that while nitrification resulted in the surface horizon (0–0.15 m) acidifying over time (Table 2), there was no evidence of acidification in the next two horizons. Movement of surface-generated acidity to deep subsoil horizons seems, therefore, to have been improbable.
Gypsum effects on percent acid saturation were strikingly evident (Fig. 2) and since acid saturation is a widely used index of lime requirement, are of particular agronomic relevance. Quite clearly, the influence of Ca and Mg movement (Fig. 1) was strongly expressed, particularly in the 0.60- to 0.75-m horizon where the relationship between acid saturation and base accumulation differentials was much better than that between acid saturation and exchangeable acidity (Fig. 1 and 2). Because exchangeable acidity differentials in the two deepest horizons were apparently strongly influenced by salt-induced acidification, they failed to accurately reflect changes that are biologically important. Patently, in terms of acid saturation, the benefits of gypsum in the 0.60- to 0.75-m horizon increased dramatically in the eighth season (1989–90) and throughout the investigation were minimal in the deepest horizon. Exchangeable acidity differentials did not mirror these effects, but similar effects were evident in terms of exchangeable Ca and Mg, extractable SO4, pHw, and pH (Fig. 1 and 2). Presumably, this surge in the ameliorative effect of gypsum resulted from above-average precipitation during the early part of the 1989–90 season (Farina et al., 2000). Since the winter months are essentially rain-free in the area of experimentation, throughflow of water would be expected to be maximal during early summer months.
Soil pH measurements (Table 1, Fig. 2) indicated that pHw provided a better measure of the ameliorative effect of gypsum than did pHs. There was a good inverse relationship between acid saturation and pHw differentials, but no meaningful relationship existed between acid saturation and pHs. Similar observations were noted in an earlier report on this study (Farina and Channon, 1988b) and these data further highlight the sensitivity of soils such as this to changes in soil solution ionic strength. The negative effects of gypsum on pHw, particularly evident in deeper horizons during early seasons of experimentation, very probably reflect the initial dominance of ionic-strength effects over those of self-liming via SO4 sorption–precipitation reactions. It is noteworthy that in the 0.60- to 0.75-m horizon, positive effects on pHw very closely matched the effects evident in terms of extractable SO4 (Fig. 1 and 2) and that below plow depth, the magnitude of the differentials in pHw were well related to the quantity of SO4 sorbed or precipitated against extraction with CaCl2. Differences between gypsum-treated and control plots in terms of pH (pHw - pHs) lend further support to the important role played by SO4 sorption–precipitation reactions. Such reactions lead to increases in cation-exchange capacity (Sumner, 1995) and the pH differentials (Fig. 2) are clearly closely related to the quantities of SO4 held against extraction with CaCl2.
|
Conclusions |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
The data reported here unquestionably support the existence of a self-liming effect as a consequence of sulfate movement. Sulfate accumulation in subsoil horizons was clearly related to appreciable increases in pHw and cation-exchange capacity, and to decreases in acid saturation. They also support previous speculation (Farina, 1997) that increases in ionic strength resulting from NO3 accumulation are capable of markedly influencing the efficacy of gypsum. While deep-subsoil acidification was not as pronounced in this study as that reported on by Farina (1997), long-term acidification in the deepest horizon (0.75–0.90 m) of both control and gypsum-treated plots without evidence of acid movement from higher horizons can only be explained in terms of NO3 accumulation, pH depression, and consequent release of interlayer Al. Apparent failure of sulfate to penetrate this horizon could conceivably have resulted from the precipitation of SO4 as an Al hydroxy–sulfate mineral (Adams and Rawajfih, 1977) not extractable with Ca(H2PO4)2. This is an aspect of gypsum use that deserves further attention. If the proposed mechanism is valid, N application rates become pivotal to the amelioration with gypsum of acid subsoils containing appreciable quantities of reserve Al.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Received for publication November 9, 1998.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Peregrina, I. Mariscal, R. Ordonez, P. Gonzalez, T. Terefe, and R. Espejo Agronomic Implications of Converter Basic Slag as a Magnesium Source on Acid Soils Soil Sci. Soc. Am. J., January 25, 2008; 72(2): 402 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.P.W. Farina, P. Channon, and G.R. Thibaud A Comparison of Strategies for Ameliorating Subsoil Acidity: I. Long-Term Growth Effects Soil Sci. Soc. Am. J., March 1, 2000; 64(2): 646 - 651. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
