| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Plant and Soil Science, Hills Bldg., Univ. of Vermont, Burlington, VT 05405-0082
Corresponding author (dross{at}zoo.uvm.edu)
Received for publication January 31, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Abbreviations: DPS, degree of phosphorus saturation • EPC0, equilibrium phosphate concentration at zero sorption • OM, organic matter • PAI, phosphorus adsorption index
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Recently, there has been a trend toward simultaneously assessing agronomic and environmental P availability. Traditional soil P tests appear related to dissolved P in runoff and/or subsurface drainage (Moore et al., 1998). Soil test P (modified Morgan's, Bray and Kurtz, Mehlich-3, and Olsen) and other extractable P forms are correlated with the quantity of P that desorbs to water, weak electrolyte extracts, EPC0 (equilibrium phosphate concentration at zero sorption), and incubated supernatant solutions (Wolf et al., 1985; Magdoff et al., 1999; Sallade and Sims, 1997b). Laboratory and field studies have shown significant relationships between soil test P quantity and soluble P concentration in field runoff, subsurface drainage, and water extracts (Daniel et al., 1993; Heckrath et al., 1995; Pote et al., 1996, 1999; Provin, 1996; Sharpley, 1995; Sharpley et al., 1996; Simard et al., 1995; Smith et al., 1995).
Although research has shown the release of phosphate to soil solutions during anoxic incubations, the actual mobilization of phosphate from the soil to overlying water has received little attention. The reduction of ferric iron (Fe3+) compounds may release occluded phosphate and ferrous iron (Fe2+) to the soil solution (Mortimer, 1941; Ponnamperuma, 1972; Holford and Patrick, 1979; Patrick and Khalid, 1974; Hill and Sawhney, 1981; Gale et al., 1992; Moore and Reddy, 1994; Sallade and Sims, 1997b). At the same time, the iron redox cycle can act as a barrier to the movement of phosphate after soil reduction. Ferrous iron may diffuse to the oxidized side of the redox interface (near the ped, soil, or sediment surface) and sorb or precipitate with phosphate when oxidized back to Fe3+ (Bartlett and James, 1995; Moore and Reddy, 1994). This mechanism becomes ineffective if the overlying floodwater becomes anoxic (Moore and Reddy, 1994) or the concentration of phosphate exceeds the capacity of Fe to reprecipitate (Lijklema, 1980).
Sallade and Sims (1997b) showed that phosphate concentrations in anoxic sediments were correlated to total Fe oxides and the DPS (0.1 M NaOH-P/P sorption capacity). They suggested that phosphate would move from drainage ditch sediments to floodwater based on the EPC0 and anoxic supernatant soluble P levels. Cooper and Gilliam (1987) have also suggested that phosphate from riparian soil mobilized to floodwater when the EPC0 was greater than the phosphate concentration of the overlying water. However, the redox potential at the sediment–water interfaces in shallow lakes and wetlands is a crucial variable with respect to internal P cycling (Moore and Reddy, 1994). Flooded field soils generally develop redox interfaces so long as the dissolved oxygen of the floodwater is not consumed. The dynamics of P cycling are completely different above and below this interface. In the anoxic zone, P is being released due to the reduction of Fe oxides and solubilization of sorbed P. Above the interface, there is the potential for P to be removed from solution through coprecipitation with or sorption on freshly precipitated Fe oxides. Thus, any laboratory experiments designed to model P transformations and mobility in wetlands that fail to account for this natural process may be inherently flawed in estimating the movement of solubilized phosphates.
Because of the potential for P release from seasonally flooded soils near Lake Champlain, we examined flooded laboratory microcosms for 14 soils from a research farm that drains into the lake. Our specific objectives for this study were to (i) determine if flooding soils of variable P fertility increases phosphate in the soil solution and floodwater over time; (ii) determine relationships between phosphate release and soil test chemical characteristics such as extractable P, Fe, and Al; and (iii) determine if Fe2+–Fe3+ transformations affect P cycling in these soils.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Site Selection and Soil Sampling
Twelve agricultural soils and two wetland soils of varying drainage, flooding regime, and fertility were sampled during a 1-wk period in October 1997 (Table 1). The nonagricultural soils were derived from the same watershed as the agricultural soils and consisted of a forested riparian wetland muck from Corbeau Creek, and an emergent marsh Lake Alice (Table 1). Surface soil samples (0–5 cm) were taken from the tops of small pits and adjacent areas. Descriptions from the Clinton County Soil Survey (Clinton County Soil Survey, 1990), coupled with visual observations, were used to assure that the soil sampled was the same series from the soils on the map. Approximately 4 kg of field-moist soil was collected at each site. Soil samples were immediately taken to the lab, thoroughly hand-mixed, and put through a 2-mm sieve. Field-moist samples were sealed in polyethylene bags and placed in a dark refrigerator at 4°C.
|
Acid ammonium oxalate–extractable Al and Fe were determined by the method of Loeppert and Inskeep (1996). After centrifugation and filtration, 4.5 mL was transferred to plastic tubes to which 0.5 mL of 1 M nitric acid was added for additional acidification. Tubes were sealed and placed in a dark refrigerator until analysis by inductively coupled plasma (ICP) using standard procedures.
Phosphate sorption isotherms were determined for each soil by treating 1 g of soil with 25 mL of solution containing 0, 0.5, 1.0, 5.0, 10.0, and 25 mg P L-1 as KH2PO4 in a matrix of 0.01 M CaCl2 (Nair et al., 1984). Duplicate solutions were placed in flasks and shaken for 24 h at 20°C. Supernatants were filtered through Whatman #2 paper and analyzed for phosphate. Isotherms were constructed by plotting the quantity of phosphate sorbed (initial phosphate added - phosphate remaining in solution after 24 h) as a function of equilibrium phosphate concentration. The EPC0 was estimated by the point where the isotherm intersected the x axis. All isotherms were inherently curvilinear. These data were log-transformed and the slope was used to measure P buffering capacity. The phosphorus adsorption index (PAI) was calculated separately by using a one-point equilibration (Bache and Williams, 1971; Mozaffari and Sims, 1994; Sallade and Sims, 1997a). One gram of soil was shaken 17 h with a 14 mg L-1 P solution (0.01 M CaCl2 as the electrolyte). The PAI was calculated by the ratio of P sorbed to log equilibrium phosphate concentration.
Redox Measurements
Reduced iron was measured using the 2,2 dipyridyl method of Muss and Mellen (1942). For the higher-P soil microcosms, the appearance of Fe(II) was only measured qualitatively. For the lower-P soil microcosms, Fe(II) was quantified by reading absorbance at 522 nm 15 min after adding reagents. Samples that contained large amounts of dissolved organics were measured on the spectrophotometer, prior to adding dye, to provide matrix correction. Soluble Fe concentrations, in selected samples, were measured by inductively coupled plasma (ICP) to confirm the need for background adjustment in the colorimetric procedure. Redox potentials of floodwater were measured weekly using an Orion (Beverly, MA) portable pH meter (Model 290A). Recorded potentials were transformed to ENHE readings by the following equation:
 |
Flooded Soil Microcosms
Between 500 and 550 mL of field-moist soil was placed in 1-L polyethylene beakers to give about a 9-cm depth of soil. The beakers (height = 
15 cm; i.d. = 9.2 cm) were gradually flooded with distilled water. Beakers were modified to permit sampling of soil solution with minimum disturbance by drilling holes (1.75 cm diam.) at approximately 5 cm from the bottom of the beakers. Plastic, stopcock-like spouts were then fitted to holes and sealed with silicone to prevent leakage. Four-micrometer mesh was affixed to the insides of the spouts to filter soil particles out of solution. All units were thoroughly washed and tested prior to adding soil. Porewater was sampled by allowing the soil solution to drain via gravity. A few milliliters of sample was allowed to drain before a sample was taken for analysis. Porewater phosphate, pH, and Fe2+ were measured about every 3 to 4 d. All samples were filtered through Whatman #1 filter paper prior to analysis. The overlying water was sampled every time the porewater phosphate was sampled. This was done by carefully inserting a 10-mL pipette midway between the soil surface and the top of the water column. A repeatable, circular-sampling motion was then used to sample floodwater. Floodwater Eh was measured weekly. All soils were studied in duplicate in a basement with no external light exposure. White fluorescent light during lab analyses was the only light soils were subjected to. The beakers were not covered and water column height was maintained by carefully adding distilled water with a syringe after every sampling. Soils were flooded in two groups; the first six soils highest in total extractable P were flooded in one group, the remaining eight soils were flooded after the first six microcosms had been studied. To minimize disturbance effects on the redox interface, soils were never stirred until after floodwater was spiked.
After soils had been inundated for 60 d (Group 1) or 85 d (Group 2), the water column was spiked with KH2PO4. Target spike phosphate concentration was 14 mg L-1; however, variability occurred (mean spike P = 12.8 mg L-1 ± 4.8). Disappearance of phosphate for the first group of field-moist soils was measured at 0, 24, 72, 144, and 264 h; the second group at 0, 43, 65, 132, 182, and 282. The temperature of the laboratory throughout the study was 22°C.
Floodwater Sampling in the Field
Floodwater of fields and stream water (Corbeau Creek) was periodically sampled during the spring and summer. Water samples were randomly taken from fields and a composite sample was then analyzed for phosphate as described previously. Samples were either immediately taken back to the lab and analyzed or refrigerated until the following morning and analyzed.
Statistical Analysis
Analysis of variance (ANOVA), linear regression, and correlation were performed using SAS (JMP) (SAS Institute, 1995). In the regressions with NH4–acetate P, the plot of the residuals deviated from normal and data were log-transformed. Before correlation analysis, data were checked for normality using the Shapiro–Wilk W Test (SAS Institute, 1995) and, because of the relatively few data points (n = 14), most data were not normally distributed. Pearson's procedure was used to correlate normally distributed data (PAI, oxalate Fe and Al, fluoride-extractable P, and P sorbed from floodwater spiking) and the coefficient is noted in the text as Pearson's r. In all other cases, Spearman's rank procedure was used. Linear contrasts and Tukey–Kramer multiple comparison tests were used for determining statistical differences between treatment means.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
In general, higher-organic-matter (OM) soils with lower P (e.g., Soils 1, 2, 3, 5, 6, and 7) showed more continuous phosphate release to the soil solution and released little of this phosphate to floodwater (Fig. 1) . Conversely, soils with higher P concentrations and lower OM (e.g., Soils 9 through 14) showed more variability in release to porewater and greater release of phosphate to floodwater (Fig. 2) .
|
|
2 mo) for all soils was 2.1 ± 0.6, and ranged from 1.0 to 3.3 mg L-1. This in itself suggests that the presence of redox interfaces enabled the large phosphate concentration gradients. As ferrous iron is mobilized to the interface region, it can be oxidized and thus re-precipitate with or sorb phosphate from floodwater. Moore and Reddy (1994) suggested that the Fe redox cycle at the interface controlled the release of phosphate to Lake Okeechobee. Porewater Eh was not measured but, instead, Fe2+ appearance was interpreted as the onset of soil reduction. A similar scenario is likely in our microcosms.
Phosphate Release and Mobility in Lower-Phosphorus Soils
Soils that had lower concentrations (
88 mg kg-1) of fluoride-extractable P displayed different phosphate release characteristics relative to higher-P soils (Fig. 1). These lower-P soils were flooded for 85 d, somewhat longer than would occur in the field. To compare with the higher-P soils, concentrations were averaged over Days 1 to 55. Mean porewater phosphate concentrations ranged from 0.043 to 0.344 mg P L-1; mean floodwater ranged from 0.015 to 0.058 mg P L-1 (Table 2). Thus, even lower-P soils had average floodwater concentrations at and above commonly reported eutrophication thresholds (0.01–0.03 mg L-1). The pH of the porewater of all microcosms increased at an average of 0.85 units over the first 10 d of flooding and then fluctuated over the remaining incubation period. These lower-P soils generally showed a steady decline in floodwater phosphate over the first 55 d (Fig. 1). The majority of the lower-P soils also shared a common effect in that decreasing floodwater phosphate was at times paralleled by increasing porewater Fe2+ and/or porewater phosphate (Fig. 1). Five out of the eight soils (1, 2, 3, 5, and 6) had porewater phosphate increases accompanied by increases in porewater Fe2+ during periods of rapid porewater phosphate release (Fig. 1). This suggests that the Fe redox cycle had important effects on the release and retention of phosphate to and from porewater and floodwater. Soils 1, 2, 3, 4, 6, and 7 reduced quickly, as Fe2+ was present in porewater by Day 15 (Fig. 1). Despite the large increases in porewater phosphate concentration in these soils, phosphate mobilization to floodwater essentially did not occur. Soil 7 (Lake Alice) had the lowest fluoride-extractable P of the soils and had the lowest porewater and floodwater phosphate concentrations (Table 1). In contrast to all other lower-P soils, Soil 8 showed a fairly continuous mobilization of phosphate to floodwater over the entire inundation (Fig 1b). Unlike other soils, Soil 8 had no decrease in floodwater phosphate at the time Fe2+ was detected (Day 34), perhaps because it had the highest release of P to the porewater of any of the lower-P soils. Most of these lower-P soils show the efficacy of redox processes in preventing solubilized phosphate from entering the water column.
|
Sallade and Sims (1997a)(b) and Cooper and Gilliam (1987) suggested that if a soil's EPC0 was greater than the phosphate of overlying water, release of phosphate to floodwater was probable and that no sorption from floodwater would occur (until EPC0 < phosphate floodwater). Theoretically, EPC0 concentrations are thought to estimate in situ soil solution phosphate concentrations. Assuming porewater phosphate concentration is similar to the EPC0, it appears that the above hypothesis is not true for the flooded soils in this study. Although porewater phosphate concentration was as much as 25 times greater than floodwater phosphate in low-P soils, virtually no movement to floodwater was observed. We hypothesize that the oxidized side of the interface adsorbed and/or re-precipitated phosphate. This has important implications for the interpretation of isotherm-derived parameters such as EPC0. Traditionally derived EPC0 measurements are performed under oxidized conditions. If intermittently flooded soils are inundated long enough to induce soil anoxia and a redox interface between floodwater and soil, it is possible that traditional EPC0 estimations do not apply for predicting phosphate mobility in wetlands. If the EPC0 is used to estimate the probability that P will move from the potentially reduced soils to overlying water (Sallade and Sims, 1997b; Cooper and Gilliam, 1987), it may need to be determined with anoxic soil and oxidized floodwater, as would be typical in nature.
Phosphate Release and Mobility in Higher-Phosphorus Soils
Average porewater phosphate concentrations (averaged over 1 to 58 d) for the 6 soils highest (>88 mg kg-1) in fluoride-extractable P ranged from 0.94 to 7.01 mg P L-1; mean floodwater ranged from 0.17 to 1.85 mg P L-1 (Table 2). Most of these higher-P soils showed fairly similar phosphate release trends (Fig. 2). All showed abrupt and substantial reductions (sixfold and about threefold) in floodwater phosphate from Days 22 to 25 despite increasing porewater phosphate intensities. Four of the soils had detectable Fe(II) concentrations beginning on Day 34 or 40. The increasing porewater phosphate and the simultaneous presence of Fe2+ suggests reductive dissolution of iron phosphates or iron oxides with sorbed phosphate (Masscheleyn et al., 1992; Gale et al., 1992; Moore and Reddy, 1994). Porewater pH rose toward neutrality over the inundation for both of these acid soils, a common occurrence in flooded soils (Ponnamperuma, 1972). The large increases in floodwater phosphate over the incubation show that reoxidation of Fe at the interface did not completely stop the mobilization of phosphate from porewater to floodwater. In general, microcosm porewater phosphate concentration appeared to increase at the time of reduction and continued to increase with time. In stark contrast, Soil 13 showed a large decrease in porewater phosphate near the time when other soils became reduced. Porewater phosphate concentration decreased about 2.3 times over three days (Days 22 to 25) and was accompanied by a 0.50-unit (7.75–8.25) increase in porewater pH and a threefold decrease in floodwater phosphate. The interaction of redox and pH probably increased reprecipitation of phosphate at the interface, causing the sudden phosphate disappearance. The high average porewater and floodwater (Table 2) phosphate concentrations for Soils 12 and 14 show that overfertilization and subsequent high extractable P have the potential to cause release of large quantities of phosphate upon reduction. Relative to other soils, 12 and 14 removed little phosphate from floodwater during the inundation (data not shown). These fields vastly exceeded the University of Vermont's (UVM) critical soil test NH4–acetate P value of 4 mg P kg-1 and also exceeded UVM's "excessive" value of 20 mg kg-1 (Jokela et al., 1998). Thus, it is statistically unlikely that this excess P will contribute to significantly greater crop yields in these fields (Jokela et al., 1998; Magdoff et al., 1999).
Soil Phosphorus Availability: Empirical Relationships
Ammonium-acetate P and fluoride-extractable P in the soils ranged from 3 to 100 mg kg-1 and 10 to 211 mg kg-1, respectively (Table 3). Three of the four soils managed in no-till agriculture had the highest NH4–acetate P concentrations (Soils 12, 13, and 14). No-till fields can accumulate P in A horizons, as the soil remains stratified from the lack of plowing (Dick, 1983; Eckert and Johnson, 1985). However, these fields may have received considerable P inputs in previous years (effects such as management and fertilizer inputs were not analyzed in this study). There was a good relationship between NH4–acetate P and the ratio of fluoride-extractable P to aluminum (r = -0.74, p < 0.003), as also found by Magdoff et al. (1999). The change in soil test P with added P has been shown to be negatively related to the quantity of NH4–acetate Al in Champlain Valley soils (Lee and Bartlett, 1977; Bartlett, 1982; Jokela et al., 1998; Magdoff et al., 1999). Magdoff et al. (1999) suggested that NH4–acetate P is an index that reflects the extent of P saturation of NH4–acetate Al and estimated the concentration of NH4–acetate P based on the ratio of fluoride-extractable P to NH4–acetate Al. Kuo (1990) also suggested that NH4–acetate P is the proportion of the P sorption capacity that is already binding P. These relationships suggest that as the number of unoccupied Fe and Al sorption sites increase, P fixation increases (P mobility decreases) and P availability decreases (Kuo, 1990; Magdoff et al., 1999). Others have shown that quantities of amorphous Al strongly affect P sorption–desorption (Harter, 1968, 1969; Cogger and Duxbury, 1984; Wolf et al., 1985; Richardson, 1985; Kuo, 1990; Bhatti et al., 1998; Lyons et al., 1998).
|
The quantity of NH4–acetate extractable P (log-transformed) explained much of the variation in DPS (r2 = 0.75, p < 0.001), suggesting that both reflect the fraction of the sorption capacity that has reacted with phosphate (Kuo, 1990; Magdoff et al., 1999). Kleinman et al. (1999) found Na-acetate extractable P to be the best single predictor of P saturation (calculated from oxalate-extractable P, Fe, and Al). Ammonium-acetate P, log-transformed, was the best single estimator (r2 = 0.85, p < 0.001) of the PAI. This suggests that the soils' capacities to sorb solution P decreased significantly as the proportion of the P sorption capacity already binding P increased (ranges given in Table 3). For both DPS and PAI, inclusion of other variables in stepwise regression did little to improve the relationships. Equilibrium-based soil P sorption was also measured by using the log-linear slope from sorption isotherms. This slope is the denominator in the Freundlich exponent and can be used to index a soil's P buffering capacity (Barrow, 1978). The P buffering capacity was best estimated by the quantity of oxalate Al (r2 = 0.58, p < 0.01) and was not related to NH4–acetate Al. The relationship with oxalate Fe was only weakly significant (p = 0.04). This again suggests that Al is affecting P sorption–desorption in these soils.
Phosphate Concentration as Affected by Phosphorus Availability
Mean porewater phosphate was highly correlated with mean floodwater phosphate (r = 0.96, p < 0.001) (ranges given in Table 2). Mean porewater phosphate concentration, as log phosphate, could be reasonably estimated from the DPS and log NH4–acetate P (Fig. 3)
. The close associations between mean phosphate concentration, DPS, and NH4–acetate P suggests that the proportion of P sorption capacity already binding P is related to the magnitude of phosphate released to porewater upon reduction. Sallade and Sims (1997b) showed that the DPS (NaOH-P/PAI) was highly correlated with phosphate in equilibrium solutions under reduced conditions. When P fractions were correlated with initial porewater phosphate and peak phosphate, relationships changed little. The DPS and NH4–acetate P were also good estimators of mean floodwater phosphate. The regression equations for estimating mean floodwater phosphate (mg L-1) based on NH4–acetate P and DPS were: log mean phosphate = 1.663 x (DPS) - 1.368 (r2 = 0.75, p < 0.01) and log mean phosphate = 1.484 x (log NH4–acetate P) - 2.278 (r2 = 79, p < 0.001). Based on the above regression equations, a mean floodwater phosphate concentration of 1.0 mg L-1 would require a DPS of 82% and a NH4–acetate P of 35 mg P kg-1.
|
Microcosm Sorption of Added Phosphate
Microcosm floodwater was spiked with P at Day 55 for the higher-P soils and on Day 84 for lower-P soils. The soils sorbed between 0% (Soil 14) and about 40% (Soil 3) of the added P. The quantity of P sorbed over the time period was negatively associated with NH4–acetate P (r = -0.78, p < 0.001) and positively associated with Alox (Pearson's r = 0.82, p < 0.001). Richardson (1985) showed that wetland soils high in extractable P and low in Alox had low PAI values. Field data showed that phosphate-saturated soils (e.g., low PAI) acted as sources of phosphate to overlying water and surrounding watersheds. The quantity of phosphate sorbed from floodwater from soil microcosms was correlated with the phosphate buffering capacity (Pearson's r = 0.78, p < 0.01) and PAI (Pearson's r = 0.76, p < 0.01). However, it was not as well correlated with DPS (r = -0.66, p = 0.016) as might be expected. These indices, measured on the preflooded soil samples, provided mixed results when used to predict P sorption under flooded conditions. However, it is clear that P additions to the floodwater can be sorbed by the soil even when there are reducing conditions within the soil profile. It is interesting that Alox showed the best correlation with this sorption, similar to findings in completely aerobic soils.
Floodwater Phosphate Concentration for Soils in the Field
Floodwater phosphate concentrations from in situ soils and water from Corbeau Creek (CC) on the 1 Apr. 1998 flooding event (soils had been inundated for about 24 h) were similar to initial microcosm floodwater phosphate concentration (Table 4). Although the phosphate concentrations were correlated, there were a few large individual differences. The higher concentration for Soil 10 in the field was probably because it was fertilized with manure during the winter (manure was present on the surface of the soil, below the floodwater, when sampled). Neither the microcosms nor the field-sampling were designed to represent entire fields. The data is presented to illustrate that short-term release of P to overlying floodwater in the microcosms was reasonably similar to that found under field conditions. Relevance of the long-term release in the microcosms is difficult to fully document without additional field data.
|
|
CONCLUSIONS AND IMPLICATIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
Agricultural soils and wetlands that receive large volumes of P inputs can become saturated with P. Unfortunately, once soils become saturated with phosphates, considerable time is required for noticeable depletion (Daniel et al., 1994). McCollum (1991) estimated that without further P additions, 8 to 10 yr of corn cropping would be necessary to reduce soil test (Mehlich III) P in a Portsmouth sand from 54 to 20 mg kg-1. Many of the agricultural regions comprising the Lake Champlain Basin are tile-drained due to poor drainage and/or closely coupled to riparian areas. Phosphate made soluble by prolonged saturation could furnish phosphate to tile drainage lines and floodwater. It is clear that soils high in NH4–acetate P and/or fluoride-extractable P can release significant quantities of phosphate to porewater and overlying water. Depending upon the soil P status, Fe availability, and soil–water redox, interfaces may act as a barrier to the mobilization of solubilized phosphates to floodwater.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONS AND IMPLICATIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. O. Young and R. D. Briggs Phosphorus Concentrations in Soil and Subsurface Water: A Field Study among Cropland and Riparian Buffers J. Environ. Qual., January 4, 2008; 37(1): 69 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Murray and D. Hesterberg Iron and Phosphate Dissolution during Abiotic Reduction of Ferrihydrite-Boehmite Mixtures Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1318 - 1327. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. I. Litaor, O. Reichmann, M. Belzer, K. Auerswald, A. Nishri, and M. Shenker Spatial Analysis of Phosphorus Sorption Capacity in a Semiarid Altered Wetland J. Environ. Qual., January 1, 2003; 32(1): 335 - 343. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Korboulewsky, S. Dupouyet, and G. Bonin Environmental Risks of Applying Sewage Sludge Compost to Vineyards: Carbon, Heavy Metals, Nitrogen, and Phosphorus Accumulation J. Environ. Qual., September 1, 2002; 31(5): 1522 - 1527. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
