| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Greenhouse and Processing Crops Research Center, Agriculture and Agri-Food Canada, Harrow, ON, Canada N0R 1G0
b Dep. of Natural Resource Sciences, Macdonald Campus, McGill Univ., 21 111 Lakeshore, Ste-Anne de Bellevue, QC, Canada H9X 3V9
c Pollution Data Branch, Environment Canada, Place Vincent Massey, 9th Floor, 351 St-Joseph Blvd., Hull, QC, Canada K1A 0H3
* Corresponding author (zhangt{at}agr.gc.ca).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: Biocarb-P, biocarbonate P • LPi, labile inorganic P • M-3 P, Mehlich-3 P • MLPi, moderately labile Pi • NaOH-P, hydroxide P • Pi, inorganic P • Po, organic P • Res-P, residual P
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In a Honeoye fine sandy loam soil, Peck et al. (1975) found that continuous applications of concentrated superphosphate over 10 yr for a total of 248 kg P ha–1 increased soil test P (Morgan's method) from 10 to 186 kg P ha–1. Halvorson and Black (1985) reported that single applications of 22, 45, 90, and 180 kg P ha–1 raised the levels of soil Olsen P by 1, 2, 4, and 8 mg P kg–1 soil, respectively, on a glacial till soil after 16 yr of P fertilization. In a 29-yr study on Ultisols in Alabama, Cope (1981) found that 24 and 20 kg P ha–1 raised Mehlich-1 soil test P by 1 mg P kg–1 when P was applied at rates of 31 and 54 kg P ha–1 yr–1, respectively. After 14 yr of annual application of P fertilizer in a Webster fine-loamy and a Canisteo fine-loamy soil located in north central Iowa, Webb et al. (1992) reported that annual P additions required to maintain Bray-1 P values increased from 16.8 to more than 33.6 kg P ha–1 with increases in the initial Bray-1 P value. Results obtained by Cox (1994) in North Carolina and Brazil showed that increases in Mehlich-3 with each unit of applied P were 0.7 units for soils with 10% clay, and it decreased exponentially to 0.2 units for soils with >50% clay. A study performed by Randall et al. (1997) in Minnesota, showed that every 1 mg P kg–1 increase in soil test P (Bray-1 P) requires 35 and 20 kg P ha–1 in a Webster clay loam soil and 58 and 26 kg P ha–1 in a Aastad clay loam soil when fertilizer P was added at 24 and 49 kg P ha–1, respectively. Clearly, the amount of fertilizer P required for each milligram of P per kilogram increase of soil test P varies with climatic conditions, soil type, soil test method employed, as well as the rate of fertilizer P applied.
In addition, without considering crop P removal, calculation of the amount of fertilizer P required for unit increase in soil test P using soil test P change and the amount of fertilizer P applied may have also caused disparities of the results. Using the net P addition approach, the difference between added fertilizer P and P removed by crops, in a study performed in a Chicot sandy clay loam soil in Quebec, Zhang et al. (1995) found that 8.6 kg P ha–1 manure and inorganic fertilizer P was required to raise soil M-3 P by 1 mg P kg–1, regardless of the inorganic fertilizer rate applied. The result was in contrast to 11.2 and 13.9 kg P ha–1 fertilizer P required for increase of 1 mg P kg–1 soil M-3 P with 44 and 132 kg P ha–1 Pi applied, respectively, when crop P removal was not deducted.
To further assess the fate of residual fertilizer P in soil, detection of various P species using selective dissolution agents can be used. A modification of a sequential extraction procedure developed by Hedley et al. (1982) fractionates soil P into various Pi and organic P (Po) forms. The fractionation procedure has significant advantages over those previously developed, such as the one for measurement of plant available P (Chauhan et al., 1981), the one for Pi (Chang and Jackson, 1957), or the one for Po (Stewart and Oades, 1972). The procedure has been used to study dynamics of P in soils ranging from slightly weathered (Hedley et al., 1982; Tiessen et al., 1983; Richards et al., 1995) to highly weathered (Ball-Coelho et al., 1993; Agbenin and Tiessen, 1994; Beck and Sanchez, 1994; Schmidt et al., 1996). In a savanna Alfisol, Agbenin and Goladi (1998) found that continuous cultivation without P fertilizer decreased the concentration of both Pi and Po. Addition of fertilizer P increased the level of LPi, but decreased the levels of Po and stable P (i.e., Res-P). Zhang and MacKenzie (1997) reported that in a Chicot sandy clay loam soil there were no changes in P fractions in soil receiving 44 kg P ha–1 yr–1 fertilizer P, but Pi and Po fractions increased in soils receiving 132 kg P ha–1 yr–1. Analysis of relationships between various Pi and Po fractions indicated that MLPi extractable with 0.1 M NaOH (NaOH-Pi) acted as the major sink for residual fertilizer P. The fraction of Res-P acted as the secondary sink for residual fertilizer P in the highland plateau soils of Ethiopia (Duffera and Robarge, 1996) and in a Chicot sandy clay loam soil of Quebec (Zhang and MacKenzie, 1997). In contrast, it was reported that moderately stable Pi extractable with 1.0 M HCl was the secondary sink for residual fertilizer P in an Alfisol (Agbenin and Goladi, 1998) and in a calcareous soil (Daroub et al., 2000). Transformations of residual fertilizer P in soil are dependent on climatic conditions, soil properties and fertilization history.
Availability of soil P to plants depends on the replenishment of soil labile P from other P fractions. A predominant fraction of residual fertilizer P in soils, NaOH-Pi, is believed to be the P absorbed by sesquioxides. This absorption may be reversible through desorption and meet crop P needs when labile P is exhausted. Guo et al. (2000) have reported that NaOH-Pi declined after biocarbonate P (Bicarb-Pi) under intensive plant uptake in eight soils varying in weathering from Vertisols and Mollisols to Ultisols and Oxisols. The study, however, was conducted in a controlled greenhouse environment. Whether the results are applicable to field conditions is not known. Using path analysis, Zhang (1996) proposed that NaOH-Pi serves as a reversible intermediate between LPi and Res-P, and the direction of the transformation depends on added P in a temperate clay soil. This hypothesis is yet to be tested.
The objectives of this study were (i) to qualify and quantify changes in soil test P using the net P addition approach, and (ii) to determine the dynamics of soil P fractions with continuous fertilization and with subsequent soil P depletion from crop removal under continuous corn in a high clay soil in eastern Canada.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Initial treatments included seven fertilizer rates with various N, P, and K combinations and two corn hybrids (A and B) at 90000 plants ha–1. The experimental design was a 7 by 2 factorial arrangement of a completely randomized block design with four replicates. Plot size remained constant at 5 m wide by 6.5 m long. The study was originally designed to evaluate the effects of fertilization on corn yield, nutrient uptake and soil nutrient build-up from 1988 to 1993. Two hybrids were included to evaluate the potential variation in the response to fertilization by different hybrids, and selected based on the annual recommendation from the local seed company. It was found that there were no significant difference between hybrids for crop yield, grain P concentration, and P uptake (Table 1). Hence, a new study was designed so that one half of the plots (continuous P addition plots) were continued with the original fertilizer regimes (i.e., hybrid A plots) and the other half (P depletion plots) which were plots previously assigned to hybrid B did not receive any additional fertilizer P and K from 1994 to 1997. Since hybrid effects were found to be insignificant in the first phase of the study and no longer of interest, the same corn hybrid was grown in these P addition and P depletion plots from 1994 to 1997.
|
Only treatments associated with fertilizer P rates were selected for this study. These treatments included long-term continuous P addition plots from 1988 to 1997 with fertilizer rates of zero (0-0-0 kg ha–1 as N-P-K), normal (170-44-141 kg ha–1 as N-P-K), and high (400-132-332 kg ha–1 as N-P-K) applied each year to the same plots, and P depletion plots from 1994 to 1997, which received the same rates of fertilizer N, P and K each year as in the corresponding continuous P addition plots from 1988 to 1993, but only fertilizer N from 1994 to 1997. In the continuous P addition plots, fertilizer P was supplied as monoammonium phosphate (NH4H2PO4) (44 and 88 kg P ha–1 in plots receiving 44 and 132 kg P ha–1, respectively) and triple super phosphate [Ca(H2PO4)2] (44 kg P ha–1 in plots receiving 132 kg P ha–1). Fertilizer P was hand broadcasted and incorporated each year shortly before planting. Sulfur and Mg were added annually at 30 kg ha–1 each in all of the fertilized plots from 1988 to 1997 to protect crop from limitation of other nutrients. All other management practices remained consistent for all plots from 1988 to 1997.
Plots were normally seeded in the first 10 d of May. Corn was seeded at an approximate depth of 5 cm using a Gaspardo planter (Model SP510, Gaspardo, Morsano AL Taglimento, Italy) with three adjustable drills. Hand-thinning to desired populations was done in mid June. Corn grains were harvested with a combine (Nurserymaster Elite, Wintersteiger America, Salt Lake City, Utah) in first 2 wk of October from two center rows in each plot. Harvest grain was weighed and grain yield determined. A subsample was retained and analyzed for P concentration, with total P removal calculated from yield and P concentration data for each plot. After harvest, the stover was chopped onto the original plots. The soil was plowed with a conventional moldboard plow to a depth of 20 cm in late October of each year. Secondary cultivation in the spring consisted of two passes of a disk harrow at a depth of 10 cm followed by one pass of a spring-tooth harrow.
Soil Sampling and Phosphorus Determination
The soil was sampled before the start of the study and each year during the experimental period, except for 1988 and 1990. One soil core with 7.5 cm i.d. and 4 additional ones with 2 cm i.d. to a depth of 20 cm were taken in each plot. Five soil cores were mixed thoroughly and a subsample taken to represent the plot. Soil samples for the analysis of soil test P were air-dried and crushed to pass a 2-mm sieve after removing visible root and crop residues. A subsample of soil was further ground to 
0.149 mm (100 mesh) for soil P fractionation.
Soil test P was measured by extracting with the Mehlich-3 solution (Mehlich, 1984), a recommended procedure for Quebec. Phosphorus in the extract was determined with a QuikChem AE Automated Ion Analyzer (Lachat Instruments, Milwaukee, WI) using the ammonium molybdate and ascorbic acid colorimetric method (Murphy and Riley, 1962). Soil test P was expressed in mg P kg–1 soil and converted to kg P ha–1 in the surface 0- to 20-cm soil depth using the bulk density of 1.24 Mg m–3 (Zhang, 1996).
Soil P in samples taken before and right after the experiment (i.e., 1988 and 1997, respectively) and before the soil P depletion (i.e.1993) was fractionated using a modified Hedley's sequential extraction procedure (Zhang and MacKenzie, 1997). Briefly, a soil sample of 0.5 g was extracted sequentially with 30 mL 0.5 M NaHCO3, 0.1 M NaOH, and 1.0 M HCl by shaking the suspension for 16 h, centrifuging for 10 min at 16000 x g, and passing through a 0.45-µm filter. A portion of NaHCO3 and NaOH extracts was acidified to precipitate extracted organic matter and the supernatant analyzed for Pi. Another portion of NaHCO3 and NaOH extracts was digested in an autoclave (103.4 kPa, 121°C for 1.0 h) with acidified ammonium persulfate [(NH4)2S2O8] oxidation and analyzed for total P. The difference between total P and Pi was considered to be Po (Tiessen and Moir, 1993). Total P in the residue after sequential extraction was determined after digestion with concentrated sulfuric acid and hydrogen peroxide. Inorganic phosphate in all extracts and digestion solutions was determined colorimetrically with the molybdate-ascorbic acid procedure (Murphy and Riley, 1962). Thus, soil P was separated into six fractions, including Bicarb-Pi and Bicarb-Po, hydroxide Pi (NaOH-Pi) and Po (NaOH-Po), acid Pi (HCl-Pi), and residue P (Res-P). Soil P fractions were expressed in mg P kg–1 soil and converted to kg P ha–1 in the surface 0- to 20-cm soil depth using the bulk density of 1.24 Mg m–3.
Data were analyzed using the Proc GLM in the Statistical Analysis System (SAS Institute, 1996). Regression analysis including both linear and quadratic models was performed using the Proc REG, but only the model with best fit was selected.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the second phase of the study (1994 to 1997), corn yield in the P depletion plots also increased with increases of the previous fertilizer rate, but decreased in comparison with the corresponding continuous fertilization plots. It appeared that the high rate of fertilizer N had to be balanced with some of fertilizer P and/or K to increase crop yield, even if the soil contained high levels of soil test P (Fig. 1) and K. Grain P concentration remained similar between plots receiving continuous fertilization and P depletion. Consequently, annual soil P removal in the P depletion plots decreased by 3.1 and 8.1 kg P ha–1 in plots previously receiving fertilizer at normal and high rates, respectively, compared with the corresponding continuous fertilization plots.
|
|
In the fertilized plots, M-3 P increased linearly over time. Compared with the initial values, M-3 P after 10 yr increased by 39 mg P kg–1 in the normal rate plots and by 157 mg P kg–1 in the high rate plots. The increases in M-3 P across all 10 yr were also linearly related to both the total amount of fertilizer P applied and the net added P. Since net P addition is the difference between the total amount of fertilizer P applied and the total crop P removals, it would be more meaningful if the net P addition is used for determination of changes in soil test P (Zhang et al., 1995). Thus, a calculated value of 6.3 kg P ha–1 of net fertilizer P application was required to increase soil M-3 P by 1.0 mg P kg–1. The value is smaller than that obtained in a Chicot sandy loam soil under the same climatic conditions, in which 14 kg P ha–1 of net added inorganic fertilizer P was required to increase soil M-3 P by 1.0 mg P kg–1 (Zhang et al., 1995). The result implies that P transformation from soluble forms derived from fertilizer to less labile forms not extractable by M-3 solution occurs slowly in soil with high clay content. The greater surface area of the clay soil means that P sorption with physically loose binding was less saturated than in the sandy soil. Reduced chemisorption of P due to large surface area has been reported for synthetic Fe oxides (Madrid and de Arambarri, 1985). Higher organic C content (32 g kg–1) in the soil under study in comparison with the Chicot sandy clay loam soil (organic C, 15 g kg–1) may have also contributed to the smaller value of fertilizer P required to increase soil M-3 P by 1 mg P kg–1 due to the increase in soil P availability by (i) organic anion replacement of H2PO–4 on adsorption sites, (ii) the coating of Fe and Al oxides by humus to form a protective cover and reduce P adsorption, and (iii) the formation of stable organic complexes with Fe and Al, preventing their reaction with H2PO–4 (Havlin et al. 1999).
Since the cessation of P fertilization, soil M-3 P in the depletion plots where fertilizer had been continuously added at the high rate for 6 yr from 1998 to 1993 declined in the subsequent 4 yr from 1994 to 1997, whereas soil M-3 P was not reduced in the depletion plots where fertilizer had been added at the normal rate from 1988 to 1993 (Fig. 1). Over the 4-yr period in the depletion plots previously receiving the high rate fertilizer, 66.4 kg P ha–1 was removed (Table 1) and soil M-3 P decreased by 15.7 mg P ka–1 (Fig. 1, regression equation). Hence, 4.2 kg P ha–1 soil P removal in grain corn resulted in a decrease of soil M-3 P by 1.0 mg P kg–1. If the M-3 P levels in the plots with zero rate fertilizer added are considered as the equilibrium values for this soil under continuous corn, with a reduction rate of 3.96 mg P kg–1 yr–1 (regression slope in Fig. 1), it would require about 28 yr to deplete the 110 mg P kg–1 of soil M-3 P built up in the plots receiving high rate fertilizer during the 6-yr period to the base-line level. The result shows an annual decrease of 3.9 mg M-3 P kg–1, which is consistent with Randall et al. (1997) who reported an annual declining rate of soil test P at 3.3 mg P kg–1 in a Webster clay loam soil in Minnesota.
Soil Phosphorus Fractions
Soil Inorganic Phosphorus Fractions
In both the continuous P fertilization and the P depletion plots, fertilizer rate, year, and their interaction significantly influenced all soil Pi fractions, except for HCl-Pi on which neither the main effects of fertilizer rate and year nor their interaction was significant (Table 2). Similarly, all effects were significant on soil total Pi in the continuous P fertilization plots, but not in the P depletion plots.
In the zero fertilizer rate plots, soil Pi fractions, including Bicarb-Pi, NaOH-Pi, and HCl-Pi, remained unchanged over time (Fig. 2 and 3) . As a result, total soil Pi remained constant over the 10-yr period from 1988 to 1997 (Fig. 4) . The results are similar to those reported previously in the same study after 6 yr of continuous corn in the same soil (Zhang, 1996), indicating that corn at the low production level, 2.6 Mg grain ha–1 yr–1 averaged across 10 yr (Table 1), was supplied from either subsurface soil P or the P released from Po fractions through mineralization.
|
|
|
With fertilizer added at the high rate, changes in soil Bicarb-Pi and NaOH-Pi followed the same patterns as those in the normal rate plots, but at greater rates (Fig. 2). Soil Bicarb-Pi increased at a rate of 15.7 mg P kg–1 yr–1 during the first 6-yr period, which was also similar to the 18.1 mg P kg–1 yr–1 in the following 4-yr period, while soil NaOH-Pi increased at a rate of 6.9 mg P kg–1 yr–1 for the first 6 yr, and at a much greater rate of 34.3 mg P kg–1 yr–1 in the following 4 yr. Thus, increased addition of fertilizers either annually from normal to high rate, or cumulatively with time, enhanced formation of NaOH-Pi, while increased fertilizer addition linearly increased Bicarb-Pi. Soil Bicarb-Pi is the fraction of P sorbed on the surface of soil particles and is believed to be labile soil Pi (Bowman and Cole, 1978). Soil NaOH-Pi is held strongly by chemisorption to Fe and Al components of soil surfaces (Ryden et al., 1977) and is considered as moderately labile soil P. The greater increasing rate of both Bicarb-Pi and NaOH-Pi with increased addition of fertilizer P can be due to the increased saturation of sorption sites on the surface of soil particles.
Soil Bicarb-Pi in the depletion plots decreased after 4 yr of continuous corn from 1994 to 1997 (Fig. 2). Compared with the value in 1993, soil Bicarb-Pi in the depletion plots that had previously received normal rate of fertilizers deceased by 6.9 mg P kg–1 in 1997, an equivalent of 17 kg P ha–1 or 7% reduction of the initial value. In the depletion plots that had previously received high rate fertilizers, soil Bicarb-Pi decreased by 50.8 mg P kg–1, an equivalent of 126 kg P ha–1 or 33% reduction of the initial value. Hence, soil Bicarb-Pi decreased at a greater rate in the plots with higher initial values than those with lower initial values.
In contrast, soil NaOH-Pi in the depletion plots, which had previously received the normal fertilizer rate remained unchanged over the 4-yr period (1993 to 1997). However, soil NaOH-Pi in the depletion plots previously receiving the high fertilizer rate increased by 19.4 mg P kg–1, compared with the value in 1993 (Fig. 2). This increase represents 83% of the decline in Bicarb-Pi after grain corn P removal was considered (Table 1). Given Bicarb-Pi is readily available to plants and had provided all the P removed in 4 yr of corn production, one may assume that the remaining portion of the decrease in Bicarb-Pi could have contributed to the formation of NaOH-Pi, because changes in other soil P fractions were found either insignificant (HCl-Pi, NaOH-Pi, Res-P) or slightly decreased (Bicarb-Po), which are to be discussed below.
Soil HCl-Pi appeared stable in the 10-yr period of corn production, regardless of the annual fertilizer rate in both the continuous P addition plots and the P depletion plots (Fig. 3, Table 1), indicating that this P fraction was in equilibrium with other fractions, or was not affected by P fertilization and crop P removals in the experimental periods. Consequently, total extractable Pi (the sum of Bicarb-Pi, NaOH-Pi, and HCl-Pi) in continuous P addition plots receiving the normal fertilizer rate increased at an annual rate of 10.1 mg P kg–1 yr–1 in the first 6-yr period and 9.2 mg P kg–1 yr–1 in the following 4-yr period (Fig. 4). In the continuous P addition plots receiving the high fertilizer rate, total extractable Pi increased at 24.3 and 48.3 mg P kg–1 yr–1 in the first 6-yr and the following 4-yr periods, respectively. This increase of soil Pi represented 57 and 64% of the added fertilizer P in normal and high rate plots, respectively. Of the increases in soil Pi fractions, 75% in normal rate plots and 49% in high rate plots was found in Bicarb-Pi, a fraction considered to be primarily available to plants (Olsen et al., 1954). In contrast, of increases in soil Pi fractions, 25% in normal rate plots and 53% in high rate plots was found in NaOH-Pi, which is considered to be slowly available to plants by desorption (Tiessen et al., 1983). These results indicate clearly that increased fertilizer P application enhanced the transformation of residual fertilizer P from Bicarb-Pi to NaOH-Pi, and consequently reduced its bioavailability. On the other hand, the storage of residual fertilizer P as NaOH-Pi may help to reduce soil P loss and thus the negative impact of soil P on surface water quality, because its stronger binding with soil components in comparison with Bicarb-Pi. However, the specific relationships between soil NaOH-Pi and P losses need to be determined. HCl-Pi is thought to represent primary mineral P, such as apatite (Williams et al. 1980), and this fraction is not readily available to plants.
No significant changes were found in soil total extractable Pi in the P depletion plots (Fig. 4).
Soil Organic and Residual Phosphorus Fractions
In the continuous P fertilization plots, fertilizer rate, year, and their interaction significantly influenced soil Po and Res-P fractions, except for total Po on which a significant fertilizer rate effect was not found (Table 2). However, in addition to Bicarb-Po, neither soil Po fractions nor Res-P was affected by the continuous P depletion with corn production.
Changes in soil Po fractions displayed different patterns between fractions and between inorganic fertilizer levels (Fig. 5) . In the zero fertilizer rate plots, Bicarb-Po remained unchanged during the 10-yr period from 1988 to 1997. Bicarb-Po was also not changed in both normal and high fertilizer rate plots during the first 6 yr (1988 to 1993), but decreased by 25 and 38% of its initial levels in the following 4-yr period (1994 to 1997) in normal and high fertilizer rate plots, respectively. Similar trend of changes in Bicarb-Po was also found in the P depletion plots where the extent of decrease in soil Bicarb-Po was greater in the plots with a previous high fertilizer rate than with a previous normal fertilizer rate. Bicarb-Po is considered as labile Po sorbed on the soil surface plus a small amount of microbial P (Bowman and Cole, 1978). This greater reduction of soil Bicarb-Po in the plots previously receiving high fertilizer rate can be due to the increased addition of fresh organic C in crop residues, which may have stimulated the mineralization of active pools of soil organic matter resulted from the boost of microbial activity. This so-called "priming effect" was also noted by Azam et al. (1993) when studying the effect of application of organic residues. This result is consistent with the findings reported by Zhang and MacKenzie (1997) in a Chicot sandy clay loam soil, and by Agbenin and Goladi (1998) in a Savanna Alfisol under continuous cultivation amended with N and P in manure. However, the results contrast with those obtained by Campbell et al. (1986) in a Black Chernozem (Mollisol) under semi-arid conditions where they found no changes in Po with added manure. The differences in Bicarb-Po dynamics among these studies may be due to the amount of added organic C and the varying climatic effects. However, the small changes in Bicarb-Po were negligible, especially in the soils containing a large amount of Bicarb-Pi, which is readily available to crops.
|
Continuous corn with zero fertilizer addition decreased the levels of soil Res-P fraction over time (Fig. 3). Addition of fertilizer at the normal rate maintained soil Res-P during the first 6-yr period, but decreased soil Res-P in the subsequent 4-yr period. However, with added fertilizer at the high rate, soil Res-P increased linearly with time at a rate of 6.2 mg P kg–1 yr–1. Soil Res-P fraction is believed to be precipitated inorganic P plus humified organic P compounds (Hedley et al., 1982). With further partitioning of Res-P, Zhang and MacKenzie (1997) showed that the majority of its variation (98%) caused by long-term fertilization is due to the presence of organic P components. Phosphorus uptake in grain corn in the zero-P plots was probably from the mineralization of Res-P and NaOH-Po. The increase of Res-P with increased fertilizer rate may have been caused by the increased addition of organic P from crop residues resulted from the higher yields (Table 1).
Regardless of the previous fertilizer rate for the P depletion plots, changes of soil NaOH-Po (Fig. 5) and Res-P (Fig. 3) were found insignificant (Table 2). As a result, changes in total soil Po (the sum of Bicarb-Po and NaOH-Po) over time were not statistically significant in any of the plots previously receiving normal and high rate fertilizer (Fig. 4), although Bicarb-Po decreased in small values (Fig. 5). As discussed previously, changes in HCl-Pi and total soil Pi (the sum of Bicarb-Pi, NaOH-Pi, and HCl-Pi) in the depletion plots were also found insignificant, while soil Bicarb-Pi decreased and NaOH-Pi increased over time. The same trend of change with Bicarb-Pi and NaOH-Pi also occurred in the plots with continuous P addition. It is hypothesized that phosphate ions physically sorbed on the soil surface and extractable with 0.5 M NaHCO3 (pH 8.5) (Bicarb-Pi) (Bowman and Cole, 1978) converted to those chemisorbed on the Fe and Al components of soil surface (NaOH-Pi) (Ryden et al., 1977; McLaughlin et al. 1977). The process was positively related to the magnitude of initial LPi in the P depletion plots or to the increases in fertilizer P rate in the continuous P addition plots.
Relationships Between Soil Test Phosphorus and Phosphorus Fractions
Soil P extractable with either M-3 solution or 0.5 M NaHCO3 is an index to the amount of soil P available to plants. It is therefore not surprising that the positive relationship was found between M-3 P and Bicarb-Pi across all the treatments (Table 3). In addition, M-3 P was significantly related to NaOH-Pi and Res-P. Interrelationships determined using path analysis has shown that NaOH-Pi is able to supply crops with available P when Bicarb-Pi is depleted due to plant uptake, and that Res-P is a supplier of the NaOH-Pi (Zhang and MacKenzie, 1997). This is confirmed with relationships among Bicarb-Pi, NaOH-Pi, and Res-P, which were discovered in this study.
|
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Residual fertilizer P converted rapidly and primarily to LPi (Bicarb-Pi) and slowly to MLPi (NaOH-Pi). It remained largely in these two fractions even over the 4-yr period after P fertilization ceased. Soil P depletion, primarily caused by crop uptake, appeared mainly from soil LPi, which was closely related to M-3 P, a recommended soil test P for Quebec. Organic P fractions were little influenced by added P, but contributed to crop available P when inorganic P was not applied. It is concluded that conversion of residual fertilizer P to relatively stable P forms in soil, such as HCl-P, is a slow process, and thus a large portion of residual fertilizer P remained available to crops in the years following application. Relationships between residual fertilizer p forms (e.g., Bicarb-Pi and NaOH-Pi) in soil and soil P losses need to be studied.
Received for publication October 7, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
