| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a ETSI Agrónomos, Universidad Politécnica, Ciudad Universitaria, 28040 Madrid, Spain
b CIFA, Alameda del Obispo, Apdo. 3092, 14080 Córdoba, Spain
* Corresponding author (rafael.espejo{at}upm.es).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Abbreviations: AAS, atomic absorption spectrophotometry • AES, atomic emission spectrophotometry • CBS, converter basic slag • ECEC, effective cation exchange capacity • ICP–AES, inductively coupled plasma atomic emission spectroscopy • L, lime • PG, phosphogypsum • Pw, water-soluble phosphorus • RG, red gypsum
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This type of soil is highly abundant in raña formations from the middle Pliocene in western Spain, where it spans more than 106 ha, and generally prevails in economically depressed areas. Rañas are detritic continental formations from the Middle to Upper Pliocene with highly weathered soils, which retain many features typical of a subtropical pre-Quaternary climate (viz. hot, wet summers, which have resulted in heavily weathered soils) (Suc, 1984; Espejo, 1987).
Until the 1940s, Xerults were minimally exploited for agricultural purposes in Spain (Espejo, 1987). In fact, they merely constituted the support for Cistus ladanifer L., Halimium ocymoides (Lam.) Willk, Chamaespartium tridentatum (L.) F. Gibbs, and Arbutus unedo L. bushes coexisting with scattered cork oaks (Quercus suber L.). The increased availability of fertilizers and scarcity of food that followed the Spanish civil war of 1936 to 1939 propitiated the clearance of the natural vegetation and the cultivation of raña surfaces. This rapidly reduced the organic matter content of the Ap horizon and also soil quality as a result. Because of their low natural fertility, the soils were used for extensive cropping of grasses (oat [Avena sativa L.] and rye [Secale cereale L.], mainly), which were cultivated under a "crop and fallow" regime (i.e., 1 yr with the crop and the next in fallow) to obtain supplementary fodder for ovine cattle in the area. In recent years, however, many crop fields have been abandoned owing to their gradual deterioration and low yields.
Rendering Ultisols suitable for agricultural purposes (particularly Xerults) entails supplying them with appropriate amendments to correct excessive acidity and Al toxicity (Peregrina et al., 2006). The effects of liming on soil Ca and Al levels are initially confined to the zone of application, the Ap horizon (Ritchey et al., 1980; Pavan et al., 1984). Because of their increased solubility, gypsum amendments applied to soil surfaces are effective at supplying Ca and lowering Al contents deep in the soil profile (Sumner et al., 1986; Sumner and Carter, 1988; Sumner, 1995; Farina et al., 2000). The use of gypsum amendments, however, not only reduces the Al content of the exchange complex, but also causes substantial leaching of exchangeable bases other than Ca, especially Mg (Reeve and Sumner, 1970; O'Brien and Sumner, 1988; Syed Omar and Sumner, 1991; Ritchey and Snuffer, 2002). Magnesium losses result from the displacement of the element by Ca and its subsequent leaching; thus, Mg ions form uncharged ion pairs with SO42– ions that quickly rise through soils with minimal sorption (Bohn et al., 1979). Because this nutrient is initially present at low levels in this type of soil, the process leads to considerably imbalanced Ca/Mg ratios (Christenson et al., 1973; Carran, 1991) that can severely impair soil productivity. In a laboratory test, Peregrina et al. (2006) found the application of gypsum at a rate of 7.5 Mg ha–1 to the Ap horizon of a Palexerult to reduce its Mg content below 2% of its effective cation exchange capacity (ECEC) after two leaching cycles. According to Adams and Henderson (1962) and Hailes et al. (1997), such a low Mg saturation severely decreases soil productivity. Therefore, gypsum-amended Ultisols require the addition of some Mg to avoid too low available Mg levels and, ultimately, Mg deficiency (Zaifnejad et al., 1996a, 1996b; Ritchey et al., 1999; Ritchey and Snuffer, 2002). Dolomite- and magnesite-based amendments have traditionally been used as Mg sources, and so has MgO obtained by heating magnesite. Thus, Ritchey and Snuffer (2002) used MgO obtained by calcination of magnesite (MgCO3) to offset Mg losses caused by the application of gypsum to Hapludults in Virginia and found that maintaining the Mg content of the exchange complex required using only half the amount of Mg when supplied as MgO than when applied as dolomite.
Using alternative soil amendments based on industrial byproducts is especially important in areas where farmers have a low purchasing power. Red gypsums are byproducts of the industrial production of TiO2 from illmenite, and phosphogypsums are byproducts of the production of H3PO4 by treating phosphate rock with H2SO4. Both byproducts have a high gypsum content and are quite affordable and accessible to local farmers to correct Al toxicity. Converter basic slag (CBS) is an industrial byproduct with a high Mg content, but scarcely used for agricultural purposes in Spain. Converter basic slag results from the demolition of the thermal insulators of converters used by the steel industry; the byproduct is rich in MgO and CaO, so it constitutes a highly efficient amendment for soil acidity (González et al., 1993).
The main purposes of this work were to assess the agronomic feasibility of using CBS as an affordable, readily accessible byproduct to offset Mg losses from gypsum-amended Palexerults in western Spain and to assess its effect on biomass production in a crop of ‘Jabato’ wheat (Triticum aestivum L.). This cultivar was previously studied in this type of soil with a view to examining the effect of lime and gypsum amendments and correcting Al toxicity, and was found to be responsive to Mg deficiency in the soil (unpublished data, 2004).
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Table 1 summarizes the properties of the studied soil profile, a clayey-skeletal, kaolinitic, acid, thermic Plinthic Palexerult (Soil Survey Staff, 2003) or hyperdistric Acrisol (FAO, 1998), as determined by Peregrina et al. (2006). The soil was examined and sampled in the center of the test field used. It has a low exchangeable base content that decreases with increasing depth—and so does its pH, while its KCl-extractable Al content exhibits the opposite trend.
|
Amendments
The gypsum amendments used to suppress Al toxicity in the soil were phosphogypsum (PG) and red gypsum (RG). In addition, a highly pure limestone (Caliza Gines) was used as a reference liming amendment. The CBS was supplied by Acerinox SA (Madrid) and obtained from a steel factory in southern Spain; the PG came from Fertiberia SA (Madrid) and the RG from Tioxide Europe SL (Huelva, Spain).
Samples were ground, sifted through 0.2-mm mesh, and desiccated at 50°C (PG and RG) or 105°C (limestone and CBS) before analysis. All were subjected to acid digestion with HF and HClO4, and to alkaline fusion with Na2CO3 (Jackson, 1976). The resulting solutions were analyzed for major, minor, and trace elements using inductively coupled plasma atomic emission spectrophotometry (ICP–AES), atomic absorption spectrophotometry (AAS), ion chromatography, selective electrodes (for F), and colorimetric methods. Phosphorus was determined according to Murphy and Riley (1962). The relative richness in CaSO4 of the PG and RG was determined according to Lagerwerff et al. (1965) by dissolving 0.5 g of the amendment in 1 L of distilled water and stirring at 25°C for 12 h—instead of 0.5 h—after which the supernatant was analyzed for Ca (AAS) and SO42– (ion chromatography).
Tables 2 and 3 list the major and minor components, respectively, of the amendments. The data for PG and RG were previously reported by Peregrina et al. (2006). The PG was the amendment containing the largest amount of gypsum; by virtue of its origin, F and P2O5 prevailed in it—and so did Fe2O3 and TiO2 in RG. The RG and PG amendments contained 82.4 and 90.7%, respectively, of CaSO4·2H2O.
|
|
The plots treated with PG + L and RG + L received the same amount of PG or RG as the previous ones; because CBS supplied additional Ca and increased soil pH (González et al., 1993), such plots received the amount of L (1.1 Mg ha–1) needed to raise the pH of the Ap horizon to the same extent as 0.9 Mg ha–1 of CBS.
Plots were prepared and amendments applied in October 2000. The amendments were surface applied and incorporated into the Ap horizon by using a harrow pass with a two-wheel tractor. In October 2001, the Ap horizon in each plot was sampled to a depth of 0.2 m at three points lying at the vertices of an equilateral triangle of 1-m side length located in the center of the plot. Then each plot was supplied with an N–P–K fertilizer consisting of 70 kg ha–1 of N, 70 kg ha–1 of P, and 70 kg ha–1 of K that were obtained from Ca-free sources (viz. KCl, NH4NO3, and NH4H2PO4, respectively). Fertilizers were applied to the soil surface and incorporated into the Ap horizon by harrowing with a two-wheel tractor. The plots were sown with Jabato wheat (Triticum aestivum L.) at a rate of 175 kg ha–1 on 5 Nov. 2001; seeds were spread on the surface and then tilled with a pass of a field cultivator. The plots were supplied with an additional 70 kg ha–1 of N in the spring. Productivity in each plot was assessed after ear maturation on 16 June 2002 by harvesting the plant material contained in two 1-m2 squares; plants were cut 2 cm above ground level. For the period 1963 to 1996, the average of the highest and lowest temperatures in January (the coldest month November–June) were 10.6 and 3.4°C, respectively, and those of June (the hottest month), 28.1 and 14.5°C, respectively.
Because the soil was too dry after plants were harvested, we postponed further sampling until the first rain event, in September—the summer in southwestern Spain is very dry—to facilitate use of the drill. On 26 Sept. 2002, the four plots per treatment were sampled similarly as in 2001, using a drill 0.1 m in diameter at three different depths, namely, 0 to 20, 30 to 45, and 55 to 70 cm, which corresponded to the Ap, AB, and Bt1 horizons, respectively.
Greenhouse Productivity Test
The Bt horizon of soils in the Cañamero raña frequently saturates with water during the rainy season (November–April). This can affect crop growth and development. In order to obtain reliable productivity data, we conducted a greenhouse study. In autumn 2001, after the fertilizer application, the 0.2-m upper layer of soil (Ap horizon) was collected from each plot by using a cylindrical drill 0.1 m in diameter. Each plot was sampled at three different points (viz. at the center and at two points 1 m on each side in the direction of the largest plot). The three samples thus obtained were mixed, homogenized, and allowed to air dry before a 1.5-kg portion was selected for placement in a 1.5-L pot 0.12 m in diameter that was previously filled with a 0.03-m-thick layer of coarse quartz sand intended to facilitate drainage. Each pot was equipped with a soil moisture sampler (Rhizon SMS-10 cm, Eijkelkamp Agricsearch, Giesbeek, the Netherlands) that was placed at a 45° angle from horizontal. The Rhizon SMS consisted of a 0.1-m-long hydrophilic porous polymer sheath (2.5-mm o.d., 1.5-mm i.d., average pore diameter 
0.1 µm) fitted around a stainless steel wire and connected to polyvinyl chloride tubing with a needle at the other end. This allowed an integral soil water sample to be collected from the 0.01- to 0.10-m depth. The soil in each pot was placed as three successive layers of 0.5 kg each, pressed by hand until they reached the same height in each pot so that all pots were filled to the same height and volume of soil. The bulk density of the soil in each pot was 1.48 Mg m–3; such a high value was a result of the Ap horizon containing 24.6% weathered quartzitic "gravel" less than 2 cm in diameter and possessing a red core and a black cortex rich in hematite and maghemite (Peregrina et al., 2006).
After the soil was taken to field capacity (0.21 kg water kg–1 soil), each pot was planted with three seeds from the same wheat cultivar (Jabato) used in the test field. Once sown, the 24 pots were placed in a randomized complete block design on a greenhouse bench, where seedling growth was monitored from November to May. The greenhouse was kept unheated through the growing cycle to ensure that the prevailing temperature would be close to that in the field, where the wheat crop is typically sown in November and harvested in June. The average of the highest and lowest temperatures in the coldest month of this period (January) were 12.8 and 6.1°C, and those of the hottest month (May) were 29.2 and 16.1°C. During the 24 wk of the greenhouse test, the pots were supplied with 150 cm3 of distilled water on a weekly basis; each water addition was done gradually to avoid the formation of free water on the pot surfaces. Such additions made the volume to 3.6 L, which, based on the sampled soil surface area (80 cm2), was equivalent to the approximately 450 mm by which precipitation exceeded evapotranspiration in the months where the former was greater than the latter (October–April). After half the volume (1.8 L) was added (viz. 12 wk after seeding), and 2 d after the last 0.15 L of distilled water was supplied, the soil solution was extracted by sticking the needles from the Rhizon SMS into vacuumed bottles, and an aliquot of 0.02 L was used for analysis.
After the test, the plants were cut 0.01 m above soil level and dried to a constant weight at 60°C to determine the total dry matter weight of each plant and to measure ear and stem lengths, ears being weighed separately. Following drying, the soil contained in each pot was saved for analysis of the exchange complex.
Analytical Methods for Soil
Soil Samples
Both the soil samples used to describe the profile and those collected from the test field and the pots were subjected to the same analyses. They were air dried and sifted through 2-mm mesh. The soil pH was determined in water and in 0.01 mol L–1 CaCl2 (using a soil/solution ratio of 1:2.5). The electrical conductivity in water was also determined at a soil/solution ratio of 1:2.5. Exchange bases were extracted with 1 mol L–1 NH4OAc at a soil/solution ratio of 1:10 (Thomas, 1982), the extracts being used to quantify Ca and Mg by flame AAS, and Na and K by flame AES. The amounts of Ca, Mg, Na, and K extracted in water at a soil/solution ratio of 1:10 were subtracted from the previously measured amounts of exchangeable bases. Aluminum was extracted with a 1 mol L–1 KCl solution (Barnhisel and Bertsch, 1982) and determined by titration with NaF (Yuan, 1959). The water-extractable P contents were determined by using the method of Sissingh (1971), except that the final soil/water ratio used was 1:5 instead of 1:22, and the Bray-1 extractable P (Bray and Kurtz, 1945) was also determined. Phosphorus in the resulting extracts was measured colorimetrically according to Murphy and Riley (1962).
Soil Solutions
The soil solutions were analyzed for Ca, Mg, K, Na, Al, Si, Zn, and Mn by ICP–AES and AAS, SO42– by capillary electrophoresis, NO3– and Cl– with ion-selective electrodes, F– also with an ion-selective electrode (using a TISAB solution to previously destroy F complexes), and P by using the colorimetric method of Murphy and Riley (1962).
The activity of the major ionic species of Al present in the soil solution was determined by using the software MINTEQA2/PRODEFA2 (Allison et al., 1991), using the concentrations of Ca, Mg, K, Na, Al, Si, Zn, Mn, SO42–, NO3–, Cl–, and F– in the soil solution extracted from each pot as inputs.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Effect of Amendments on the Soil Exchange Complex and Phosphorus Availability
Ap Horizon in Plots
One year after the amendments were applied, there were no significant differences with respect to the control as regards pH in soil/water suspensions in the Ap horizon, except that the L-treated samples had an increased pH (6.4) (Table 4
). On the other hand, the pH in CaCl2 was significantly increased with respect to the control for all treatments; this is consistent with the Al content on the exchange complex, which was always significantly lower than in the controls as a result of Al being displaced by Ca. Two years after application, all gypsum treatments had caused a small—but significant—increase in pH with respect to the control; the increase was especially marked in the L-treated plots, which exhibited significant differences from the gypsum-treated plots. The pH in CaCl2 changed similarly as in the previous year; this is consistent with a gradual decrease in the amount of Al present in the Ap horizon of the amended plots.
|
The initial Mg content of the soil was very low: 0.22 cmolc kg–1, and also a low percentage of the ECEC was saturated by Mg (8.8%) (Table 4). Hailes et al. (1997) studied Udults and other types of acid soils in northeastern Australia and found exchangeable Mg contents of 0.21 cmolc kg–1 and an ECEC saturation by Mg of 7% as the threshold level for a favorable response of maize (Zea mays L.) to the addition of Mg to the soil.
One year after our amendments were applied to the plots treated with PG + CBS or RG + CBS, the exchangeable Mg contents were significantly higher than those of the controls or the plots that were supplied with the other amendments (Table 4). After 2 yr, the gypsum treatments containing CBS continued to significantly increase Mg contents with respect to the L-based treatments; the differences with respect to the control were only significant with the PG + CBS treatment, however. The increased Mg content of the L treatment relative to PG + L and RG + L treatments must have been the result of a much higher Ca/Mg ratio in the soil solution of the plots that received gypsum in addition to limestone and of the increased negative charge in the plots treated with only limestone—the charge being offset by the effect of Ca not displacing Mg from the exchange complex. Magnesium in the PG + L and RG + L treatments saturated the ECEC by 4 and 3.7%, respectively; by contrast, Mg in the control saturated it by 9%. In a study on Mg availability in acid soils (Udults mainly) from Alabama, Adams and Henderson (1962) established 4% saturation of the exchange complex by Mg as the threshold for Mg deficiency in the soil leading to Mg deficiency in the resulting forage as well.
According to Zaifnejad et al. (1996a, 1996b), applying gypsum-rich industrial byproducts from flue gas desulfuration to Typic Hapludults increased the exchangeable Ca contents and decreased Al, but elicited no favorable response with respect to the control in an ‘Artur’ wheat crop in biomass production or chlorophyll in the leaves, which they ascribed to a decreased availability of Mg. In further work on the same soil, Zaifnejad et al. (1996a) and Ritchey et al. (1999) found wheat cultivated in Ultisols amended with gypsum alone to exhibit severe symptoms of Mg deficiency and concluded that avoiding hypomagnesemia in cattle would require adding some Mg to the gypsum amendment. Magnesium deficiency in wheat forage can be worsened by a low availability of P in the soil (Reinbott and Blevins, 1994) and also by excessive soil moisture (Karlen et al., 1980); these two conditions are prevalent in our soils in autumn and winter. Based on the foregoing, the Mg levels on the soil exchange complex in the plots amended with PG + L or RG + L can result in a risk of hypomagnesemia in cattle fed with the crop, especially as a result of the increasing loss of Mg with time.
One year after the amendments were applied, the exchangeable Na content had been reduced in all treatments (Table 4). The differences, however, were only significant for the RG + CBS treatment. After 2 yr, the Na content was significantly decreased with respect to the controls for all treatments, the effect being more marked for the gypsum-treated than the only L-treated soils. The exchangeable K contents exhibited no significant differences from the control with any treatment 1 yr after application; after 2 yr, treatments PG + L and RG + L resulted in significantly lower contents relative to the controls, however. The disparate dynamics of K relative to Mg and Na can be partly ascribed to the fertilizer supplying K and to the specificity of K adsorption sites in illites and vermiculites (Syed-Omar and Sumner, 1991). As noted above, the 1 mol L–1 KCl-extractable Al content was reduced with respect to the control for all treatments (particularly for the L-treated plots).
The water-soluble P (Pw) contents 1 and 2 yr after application of the amendments were significantly higher in the L-treated plots than in the controls (Table 4), which can be ascribed to the pH increase in the former. In these soils, phosphate adsorption at pH 4.1 to 6.5 is strongly influenced by the pH—it decreases markedly with increasing pH (Espejo and Cox, 1992). With the other treatments, Pw contents were lower than those of the controls, albeit not significantly, possibly as a result of the precipitation of complex phosphates of Al and Ca or even, as suggested by Phillips et al. (2000), the precipitation of dicalcium phosphate or dicalcium phosphate dehydrate. In Bray-1 P, no significant differences between treatments were observed. Bray-1 P contents in uncultivated Palexerults under cork oak or bushes were in the region of 5 to7 mg kg–1 (i.e., very low) (Espejo, 1993).
AB Horizon in Plots
The main difference in changes in the pH in soil/water suspensions in the AB horizon with respect to the Ap horizon was that neither CBS nor L was as effective in increasing pH in the former (Table 5
). Two years after the soil was amended, all gypsum-treated plots exhibited a pH in water that was significantly lower than that of the AB horizon of the controls. This can be ascribed to the salt effect caused by gypsum in the percolation water received from the Ap horizon. The origin of the decreased pH must have been Al on the exchange complex being displaced by Ca from the gypsum. The pH in CaCl2 was not subject to these differences, and all gypsum treatments resulted in a significantly increased pH relative to the controls as an effect of their containing low levels of exchangeable Al (see below).
|
The proportion of 1 mol L–1 KCl-extractable Al was significantly reduced by all gypsum treatments (Table 5); with the L treatment, the reduction was somewhat lower—albeit not significantly—as a result of the limestone being less soluble than gypsum.
The Pw and Bray-1 P contents were very low, with no significant differences between treatments. This was a result of the decreased pH and increased clay and Fe hydrous oxide contents, which increased P adsorption to a great extent (Espejo and Cox, 1992).
Bt Horizon in Plots
The Bt1 horizon in the gypsum-treated plots exhibited the same trends as the AB horizon with regard to pH in soil/water and CaCl2 suspensions (Table 5). Thus, the gypsum amendments increased the exchangeable Ca contents and decreased the Al contents relative to the controls. The Mg levels were significantly increased in all CBS-treated plots. The increase was, in part, due to Mg being displaced from the Ap and AB horizons, as seen in the RG + L and PG + L plots. Similar results in this respect were previously obtained by Sumner et al. (1986). Neither Na nor K changed significantly with respect to the controls, however, partly as a result of the Na received from the Ap and AB horizons.
The Pw and Bray-1 P contents were both very low (in the case of Pw, below the detection limit) with no significant differences between treatments, as in the AB horizon (Table 5). Also as in the AB case, the very low P availability is the result of the low pH and increased clay and free Fe hydrous oxide contents.
Ap Horizon in Pots
Soil in the pots exhibited similar—albeit more marked—trends in pH, electrical conductivity, and exchangeable bases and Al to those of the Ap horizon in the plots 2 yr after the amendments were applied (Table 5). The differences in Ca contents between treatments persisted, but contents were slightly lower in the pots than in the corresponding plots. The PG + L and RG + L treatments caused the greatest losses of Mg from the exchange complex, and so did all gypsum treatments as regards Na; therefore, the Ca/Mg ratio in the pots containing PG + L soil was 20 (vs. 4 in the control) and the ECEC saturation by Mg was 3.2% with PG + L and 2.8% with RG + L (vs. 8.8% in the control). The differences in the evolution of the exchangeable base contents between the Ap horizon in the pots and in the plots can be ascribed to better drainage and leaching conditions of the former, and to poor drainage in the field. Also, a portion of the salts lost through drainage during the rainy season under field conditions returned to the soil by ascending capillary flow from the deeper horizons during the summer (a rainless season with heavy evapotranspiration). As noted above, raña deposits lie on impermeable sediments from the Miocene; this facilitates the formation of perched water layers that rise to the outside very slowly. The Mg content of the pots containing CBS-treated soil from the Ap horizon continued to be significantly higher than those containing Ap horizon soil treated with PG + L or RG + L.
The Pw and Bray-1 P content exhibited the same tendencies as those in the samples from the Ap horizon in plots 2 yr after the amendment application (Table 5).
Effect of the Amendments on the Composition of Soil Solution in Pots
The Mg content of the soil solution in the pots containing Ap horizon soil amended with PG + L or RG + L was lower—albeit not significantly—than the controls (Table 6
); however, it was reduced to half. On the other hand, the Ca content was significantly higher (up to about five times); the high Ca concentration in the soil solution due to gypsum caused Mg to be displaced from the exchange complex and removed in the percolation water as MgSO4; as a result, the Ca/Mg ratio was 10 times higher in the soil solution from the pots with soil treated with PG + L and RG + L than in the controls. The Mg content of the soil solution from the pots containing L- treated Ap horizon soil was not significantly different from that of the control; this can be ascribed, as discussed above, to a less marked loss of Mg from the exchange complex as a result of the lower Ca concentration (with respect to gypsum-treated soil) in the soil solution and an increase in the negative charge on the exchange complex.
|
The F content of the soil solution from the pots with Ap horizon soil from the gypsum-treated plots was increased with respect to the controls (Table 6) due to the F present in the natural soil, which was displaced by SO4 (Peregrina et al., 2007). In these soils, the F content derives from previously applied phosphate rock fertilizer (Peregrina et al., 2006). The increase was maximal in the pots filled with soil from the plots treated with PG, which contains 12.3 g kg–1 of F (Table 2); the increase in F was accompanied by significant differences in Si contents as a result of F dissolving silicates (Arocena et al., 1995). The lowest F contents were those of the pots filled with soil from the plots that where amended with only L, where the pH increase facilitated precipitation of CaF2 (Larsen and Widdowson, 1971; Street and Elwaly, 1983; Mackowiak et al., 2003; Ruan et al., 2004) and F adsorption in the soil matrix: a pH increase favors precipitation of Al amorphous hydroxides, which possess a high F– adsorption capacity (Omueti and Jones, 1977; Ruan et al., 2004). In Palexerults similar to ours, Garrido et al. (2003) found the formation and precipitation of Al hydroxypolymers to be the main source of Al toxicity alleviation following application of liming amendments (Wenzel and Blum, 1992).
The activity of the most rhizotoxic mononuclear Al3+ species (Parker et al., 1989; Kinraide, 1991; Kochian, 1995) was reduced with respect to controls by all treatments (Table 6). Differences were not significant with the RG + L treatment, however, the lowest levels being those in the soil solution from the pots filled with L-treated soil; as noted above, this can be a result of the pH increase favoring precipitation of Al as insoluble hydroxides. The Al3+ activity was also very low with respect to the control in the soil solution from the pots filled with PG-treated soil due to the high affinity of Al for F– (Martin, 1996); because, as noted above, all gypsum-treated soils exhibited increased F contents relative to the controls, such affinity prevented the gypsum treatments from increasing the activity of Al–SO4 pairs relative to the controls.
Effect of Converter Basic Slag on Aerial Biomass Production
Field Tests
Despite the large differences in biomass production, only the plots treated with RG + CBS exhibited a significantly increased output with respect to the control (Fig. 1
). This was a result of the high precipitation in the Cañamero raña during the period November 2001 to April 2002 (viz. 1695 mm vs. an annual mean of 830 mm for the period 1963–1996). The excess precipitation affected some areas in the test field more markedly than others and resulted in large production differences between plots in each block. As a result, while the RG-treated plots were more productive than the PG-treated plots, differences were not statistically significant at the 0.05 probability level.
|
|
In previous work, Peregrina et al. (2006) ascribed the decreased productivity of PG-treated soils relative to RG-treated soils to the increased Zn content of RG; in fact, this type of soil contains very little Zn (Santano et al., 1993); however, the Zn contents of the soil solutions obtained in this work exhibited no significant differences between treatments (Table 6). Based on our results, the observed production differences may have resulted from an increased activity of Al–F ion pairs in the soil solution from the Ap horizon in the PG-treated plots. These ion pairs were formerly deemed nontoxic (Cameron et al., 1986; Wright et al., 1987; Tanaka et al., 1987; Noble et al., 1988; MacLean et al., 1992; Kinraide, 1991), possibly because the soil solutions examined at the time contained them in very low proportions (Kinraide, 1997). Subsequent tests exposed a potential toxic effect in species such as AlFx(3–x), AlF2+ and AlF2+ (Stevens et al., 1997; Kinraide, 1997), due to their possible interference with P uptake by plants (Facanha and de Meis, 1995; Facanha and Okorokova-Facanha, 2002). The biomass production differences between the PG + L and L treatments, and the MINTEQ-predicted ionic activities of Al–F ion pairs in the soil solution from the Ap horizons that received such treatments, are consistent with this assumption.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Received for publication September 6, 2006.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  I. Mariscal-Sancho, R. Espejo, and F. Peregrina Potentially Toxic Effects of Phosphogypsum on Palexerults in Western Spain Soil Sci. Soc. Am. J., January 21, 2009; 73(1): 146 - 153. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
