| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Crop, Soil, and Environmental Sciences, Univ. of Arkansas, 115 Plant Sciences Building, Fayetteville, AR 72701
b Dep. of Soil Science, Univ. of Wisconsin, 1525 Observatory Dr., Madison, WI 53706-1299
c Environmental Resources Center, Univ. of Wisconsin, 1450 Linden Dr., Madison, WI 53706
* Corresponding author (kbrye{at}uark.edu)
Received for publication February 9, 2001.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Abbreviations: CP, chisel-plowed • ETL, equilibrium-tension lysimeter • f, fertilized • MRP, molybdate-reactive phosphorus • nf, unfertilized • NT, no tillage • OM, organic matter • TDP, total dissolved phosphorus
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
Historically, most data on P movement in soil have been based on soil analysis of extractable P as a function of depth, which has led to the general assumption that no substantial vertical P movement or leaching loss occurs because of high P-fixation capacity in many mineral soils (Heckrath et al., 1995; Sims et al., 1998; Sui et al., 1999). Soil erosion caused by surface runoff has typically been the primary mechanism of P loss from soil to receiving waters. Significant P leaching can occur where certain combinations of land-use practices (i.e., overfertilization and/or excessive manuring), soil properties (i.e., sandy subsoil, high organic matter, and the presence of preferential flow paths), and climatic conditions (i.e., precipitation > evapotranspiration) exist (Eghball et al., 1996; Sims et al., 1998). Nonetheless, subsoil leaching of P has generally been considered insignificant and unimportant from agronomic and environmental points of view (Heckrath et al., 1995; Sims et al., 1998; Hesketh and Brookes, 2000). However, recent field studies have indicated the contrary (Table 1), suggesting that soil solution concentrations and subsurface P leaching losses are larger than once thought (Heckrath et al., 1995; Sims et al., 1998; Hooda et al., 1999).
|
Tension lysimeters measure unsaturated water flow, rather than only saturated water flow, which is measured by zero-tension lysimeters or tile drains. Tension lysimeters have a known area through which leachate solution flows and is collected (i.e., the area of the porous plate), whereas tile drains generally do not; thus a contributing area has to be assumed before flux calculations can be made, causing estimates of leaching losses to be suspect. However, in the case of tension or tensionless lysimeters and tile drains, the contributing volume of soil sampled is not exactly known. The use of tension lysimeters also avoids anaerobic conditions due to ponding above the lysimeters. This feature becomes important with P studies because P adsorption in soil decreases under prolonged saturation as a result of the reducing environment, which causes P solubility to increase (Sah and Mikkelsen, 1986). Additionally, more natural unsaturated water flow patterns are maintained with equilibrium- than with fixed-tension lysimeters.
Much of the recent literature has identified the need for direct field measurements of P leaching as a high research priority (Mozaffari and Sims, 1994; Heckrath et al., 1995; Sims et al., 1998; Hooda et al., 1999; Hesketh and Brookes, 2000). Seasonal responses to tillage, crop rotations, method and timing of P fertilization, and manuring must be a research focus in the future if P loss from agricultural soil to sensitive aquatic ecosystems is to be minimized (Sims et al., 1998). One question that needs to be addressed is, What are the P concentrations and P loads that leach from various natural and managed ecosystems and potentially enter surface waters through base flow?
We hypothesized that (i) a low level of P leaches from natural and managed ecosystems, (ii) P leaching is greater from managed corn systems than from a natural restored tallgrass prairie, (iii) P leaching is greater from chisel-plowed than no-tillage corn systems, and (iv) P concentrations in leachate solutions are correlated to water-extractable soil P concentrations at depth in the soil below the root zone. We used equilibrium-tension lysimeters to evaluate and compare P leaching among natural (i.e., a restored tallgrass prairie) and managed ecosystems (i.e., N-fertilized or N-unfertilized and no-tillage or chisel-plowed corn agroecosystems) in a fine-textured silt loam soil in southcentral Wisconsin.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
A randomized complete-block design with four replications was established for conventional chisel-plowed (CP) and no-tillage (NT) corn treatments in fall 1994 (Brye et al., 2000). A 105-d relative maturity hybrid corn variety was planted for each tillage treatment at two fertilizer rates representing optimal and deficient N requirements for corn (Kelling et al., 1998). Fertilized (f) tillage treatment combinations received 180 kg N ha-1 yr-1 of surface broadcast-applied ammonium nitrate immediately following planting, while the N-unfertilized (nf) tillage treatments received no supplemental N. Starter fertilizer (10 kg P ha-1 and 25 kg K ha-1) was banded 5 cm laterally and 5 cm below the corn seed at the time of planting in all corn treatments. The corn plots were last limed (i.e., 4.5 Mg ha-1) in November 1989 and manure was last applied in October 1992 (i.e., 34 Mg ha-1 of sheep manure). Corn grain was harvested annually and residue returned to the field in the corn systems. For the chisel-plow treatment, tillage occurred in the fall following harvest and the seed bed was prepared using a disk in the spring before planting.
Four 7- x 7-m plots were established at the prairie site in spring 1995 (Brye et al., 2000). The prairie had been restored in June 1976 after >30 yr of agriculture; the current vegetation is classified as mesic tallgrass prairie. The prairie was last burned on 18 Apr. 1998. A more detailed description of the sites can be found in Brye (1997) and Wagai et al. (1998).
Soil Organic Matter, pH, and Extractable Phosphorus
Soil samples were collected at 30-cm depth increments to 1.2 m before planting in April 1995 and at 0- to 5-, 5- to 15-, 15- to 30-, 30- to 60-, 60- to 90-, and 90- to 120-cm depth increments in November 2000 from each of the four plot replications in the prairie and corn agroecosystems. Two 2.0-cm-diameter soil cores were collected per plot and mixed to produce a single composite sample per plot. Soil samples were dried for 48 h in a forced-draft soil dryer at 33°C and ground to pass through a 2-mm mesh screen. Soil organic matter (OM) was determined by loss on ignition and soil pH was determined for a 1:1 soil and water paste. Soil P tests included extraction using distilled water (Kuo, 1996) and Bray–Kurtz P1 (Frank et al., 1998) using 2.5 g of dry soil and 25 mL of extractant. Phosphorus in the extracts obtained by each method was determined colorimetrically by the ascorbic acid method (Murphy and Riley, 1962) using a Beckman DU 640 spectrophotometer (Beckman Coulter, Fullerton, CA). Water-extractable soil P was reported as mg P L-1, while Bray-extractable soil P was reported as mg P kg-1 of oven-dry soil.
Precipitation
Hourly precipitation was measured on-site at the prairie using a tipping bucket rain gauge from June 1995 through December 2000. Similarly, a micrometeorological weather station located <150 m from the agricultural site recorded hourly precipitation with a tipping bucket rain gauge. Thirty-year mean monthly precipitation records for the Arlington Agricultural Research Station were also used for comparison (Owenby and Ezell, 1992).
Drainage and Phosphorus Leaching
Equilibrium-tension lysimeters (ETLs; 0.76 m long x 0.25 m wide x 0.13 m deep) were used to monitor drainage from the restored prairie and N-fertilized and N-unfertilized NT and CP corn agroecosystems (Brye et al., 1999). Equilibrium-tension lysimeters measure water flow (i.e., drainage) and solute transport below an undisturbed soil column. Replicate 0.2-µm, porous, stainless-steel ETLs (0.19 m2; with a porous plate bubbling pressure of 66 kPa) were installed at 1.4 m below the soil surface in the restored prairie and N-fertilized tillage treatments during fall 1995. Replicate ETLs were installed at similar depths in the N-unfertilized tillage treatments during fall 1999. All ETLs were acid-washed before installation. Heat dissipation sensors were placed immediately above the porous plate of each ETL and in the surrounding bulk soil to continuously monitor the matric potential at the two locations (Reece, 1996; Brye et al., 1999). A portable, regulated vacuum system provided continuous suction to the porous plate of the ETLs (Brye et al., 1999). The regulated vacuum system was adjusted manually several times a week to provide suction that was slightly more negative (i.e., a few kPa) than the matric potential recorded in the surrounding bulk soil with the heat dissipation sensors. Brye et al. (1999) contained schematic diagrams of the ETLs and vacuum system and a cross-sectional view of the final installation of an ETL with heat dissipation sensor placement.
The ETLs were sampled under vacuum every 2 wk between March and December and once every 4 wk during the rest of the year (Brye et al., 1999). Leachate was collected from the ETL's collection reservoir, which can contain 23 L (i.e., equivalent to 110 mm) of water, through a high-density polyethylene tube that was inserted into a stainless steel sampling tube that extends from the drain port of the lysimeter to the soil surface. The initial leachate (up to 1 L) was collected into a 1-L high-density polyethylene bottle and transported back to the laboratory where the leachate volume was measured. Any remaining leachate from the lysimeters (>1 L) was collected, volumes were recorded, and the leachate was discarded. Aliquots of the initial 1 L of leachate were filtered through glass fiber filter paper (Whatman [Maidstone, UK] G6) and stored in high-density polyethylene bottles at 4°C for chemical analysis. Aliquots of unfiltered leachate were also stored at 4°C for chemical analysis.
Leachate samples collected during a 21-mo continuous period between 17 Dec. 1998 and 5 Sept. 2000 were analyzed for molybdate-reactive phosphorus (MRP) using the ascorbic acid color development method (Murphy and Riley, 1962). Leachate samples collected from January through September 2000 were also analyzed for total dissolved phosphorus (TDP) using the ammonium persulfate and sulfuric acid digestion method (USEPA, 1993). Additionally, leachate samples collected from March through April of 1996, 1997, and 1998, typically the period of the year with the largest drainage fraction of annual precipitation (Brye et al., 1999, 2000), were also analyzed for MRP as a check on concentrations measured in the 21-mo sampling period. All leachate-MRP analyses were conducted on a Beckman DU 640 spectrophotometer using 21 mL of leachate for color development. Molybdate-reactive P and TDP are reported as volume-weighted concentrations based on the cumulative mass of MRP and TDP leached and the cumulative amount of drainage measured during the study period. Mobile P (i.e., MRP and TDP) leaching losses (i.e., load) were calculated from measured drainage fluxes and P concentrations.
Statistical Analyses
Comparative studies such as ours cannot be used to illustrate cause and effect, but do provide the opportunity to deduce the effect of land use (i.e., natural prairie versus managed agroecosystems) (Wagai et al., 1998). To infer a land-use effect on soil properties and processes using a comparative study it is critical that initial site and soil conditions were similar. Historically, the restored prairie was subject to tillage and fertilization practices similar to those used on the adjacent Arlington Agricultural Research Station. Numerous lines of evidence, such as similar physiography, climate, soil type, and soil physical and chemical characteristics, suggest that our underlying assumption (i.e., the prairie and corn agroecosystems' microclimates and numerous soil conditions were similar at the start of the study) is reasonable (Wagai et al., 1998) (Table 2). Therefore, a randomized experimental design was used to compare measurements of soil properties and processes among the prairie and corn agroecosystems. This analysis assumed that the variability among the replicate plots of each ecosystem treatment was representative of the variability among plots in general that share similar soil, ecological, and climate conditions (Wagai et al., 1998).
|
= 0.05) were determined for drainage and mobile P concentrations and loading (SAS Institute, 1992) along with coefficients of variation (CVs). Pearson linear correlations were determined for the relationship between water- and Bray-extractable soil P (Minitab, 1997). Linear correlations were also determined for the relationships between mean volume-weighted MRP concentration and cumulative MRP load from 25 Apr. through 21 July 2000 and soil OM, pH, and water- and Bray-extractable soil P from November 2000.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
|
Soil test P levels were greater in the corn than prairie due to past P fertilizer (i.e., as inorganic or organic) additions. Similarly, soil test P levels below the plow layer (i.e., >30 cm) in the three ecosystems indicated little to no vertical movement of P in the soil profile (Fig. 1). Consequently, differences in soil P between the prairie and corn agroecosystems exist near the soil surface. Bray-extractable soil test P in the top 15 cm of soil was in the "high" (i.e., 21–30 mg L-1) and "excessively high" (i.e., >30 mg L-1) categories for corn production in Wisconsin at the prairie and agricultural sites, respectively (Kelling et al., 1998). High background soil test P levels in the prairie are probably due to its agricultural history prior to restoration and the limited removal of P since restoration. There was little difference in soil test P, pH, and OM between 1995 and 2000. Mean annual aboveground biomass production was greater for the corn agroecosystems, which was 12.2 (standard error = 0.5), 13.1 (0.4), 19.1 (0.7), and 21.0 (0.5) Mg ha-1 yr-1 for the N-unfertilized NT and CP and N-fertilized NT and CP treatments, respectively, than the prairie, which was 1.8 (0.2) Mg ha-1 yr-1 (Brye, 1999).
Precipitation and Drainage
The prairie and agricultural sites received the same amount of precipitation during the peak drainage months in 1996, 1997, and 1998 (Table 4). However, in 1996 and 1997, precipitation was roughly half of the 30-yr mean for March through April, while in 1998 precipitation was 75% higher than the 30-yr mean. Cumulative measured precipitation during the 21-mo continuous drainage and P-leaching monitoring period (i.e., January 1999 through September 2000) was 15 and 8% higher than the 30-yr mean for the agricultural and prairie sites, respectively. Similarly, measured precipitation during the 9-mo period (i.e., January through September 2000) was 40 and 23% higher than the 30-yr mean at the agricultural and prairie sites, respectively.
|
|
|
|
Corn root growth (i.e., rooting depth and density) has been shown to be higher under N-fertilized than N-unfertilized fine-textured, silt loam soils (Fehrenbacher and Snider, 1954). As coarse corn roots decompose annually, large root channels or macropores develop that extend from the surface deep into the soil profile. In contrast, root growth under tallgrass prairie consists of a shallower rooting depth, though higher root density near the surface, and a smaller fraction of coarse roots than under corn (Brye, 1999). Consequently, the likelihood of preferential flow, via macropore flow through decomposing root channels circumventing much of the soil matrix within the root zone, is greater under corn than prairie. Similarly, there is greater chance that roots could extend to and spread out on the surface of the ETLs in search of water in the corn than prairie, which further increases the likelihood of greater preferential flow under corn than prairie.
Significant differences in rooting depth or density due to tillage have not been observed for the corn agroecosystems of this study (Brye, 1999). On one hand, one could argue that decaying root and worm channels that remain intact from the surface down into the soil profile under NT would result in greater macropore development and water infiltration. On the other hand, one could argue that the compaction that occurs in time at the surface of a NT system would decrease macropore volume in the surface layer and its infiltration capacity compared with a CP system. Nonetheless, insufficient evidence exists for the systems in the study to conclude there is a tillage effect on drainage and P leaching due to differences in rooting enhancing macropore development and infiltration among NT and CP systems.
Leachate Phosphorus Concentrations
Mean leachate-MRP concentrations were consistently higher for the NTf corn agroecosystem compared with the prairie and CPf corn agroecosystem during the January 1999 through September 2000 monitoring period. However, non-flow-weighed MRP concentrations for the prairie and corn agroecosystems sometimes approached the detection limit (approximately 0.02 mg L-1) for the Murphy–Riley colorimetric method (Randall et al., 2000). Leachate volume-weighted MRP concentrations were 0.02, 0.09, and 0.05 mg MRP L-1 for the prairie, NTf, and CPf corn agroecosystems, respectively (Table 6), but ecosystem differences were not significantly different at the 5% level (p > 0.11) due to the high variability of P concentrations (CV = 68%). However, even though measurement differences were not significant at the 5% level, assessments of statistical significance at this level may be too stringent for data collected in lysimeter studies. A more conservative statistical significance threshold (e.g., 10%) may be more appropriate for this type of field work. Regardless, measurement precision could still be improved with greater replication of lysimeters in field plots.
Drainage between March and April accounts for a large fraction of annual drainage from the prairie and corn agroecosystems (Brye et al., 2000). Therefore, interannual variations of MRP concentrations in leachate during the largest flux period of the year were checked for consistency. Patterns similar to those observed from March through April 1999 and 2000 were also observed from March through April 1996, 1997, and 1998. Mean leachate MRP concentrations from March through April 1996, 1997, and 1998 were also consistently higher, though not significantly, in the NTf corn than in the prairie or CPf corn agroecosystems (Table 5). Similarly, tillage did not affect MRP concentrations in the corn agroecosystems. The range of mean leachate MRP concentrations from March through April 1996, 1997, and 1998 was smaller, but within the range of mean leachate MRP concentrations measured from March through April 1999 and 2000 (i.e., <0.01 to 0.07 mg MRP L-1).
Volume-weighted MRP and TDP concentrations were similar and unaffected by ecosystem (i.e., prairie versus agriculture) or tillage (i.e., no tillage versus chisel plow) during the January through September 2000 sampling period (Table 7). However, volume-weighted MRP and TDP concentrations were significantly higher (p < 0.01) in the N-fertilized corn than in the prairie or N-unfertilized corn treatments even though all corn treatments received the same P-fertilizer additions. On a per sample date basis, leachate-P concentrations ranged from 0.02 to 0.09 mg MRP or TDP L-1 for the N-fertilized corn agroecosystems and from <0.01 to 0.02 mg MRP or TDP L-1 for the prairie or N-unfertilized corn agroecosystems. The mean MRP fraction of TDP in ETL leachate solutions ranged from 0.76 to 0.91 for the prairie and corn agroecosystems, which demonstrates that a small amount of soluble P is in the organic form rather than solely inorganic (i.e., orthophosphate).
Since the mobile P in the leachate solutions could be coming from anywhere within the 1.4-m soil profile or from the 2 to 5 cm immediately above the ETL surface, several possible explanations exist to account for higher P concentrations in leachate solutions under N-fertilized than N-unfertilized soils cropped to corn despite the same fertilizer P additions to all corn treatments. One possibility is that the broadcast nature of applied fertilizer N may have influenced the ionic strength of the soil solution causing a shift in the soil–soil solution P equilibrium. Ojala et al. (1983) demonstrated that short-term acidification at the surface, as a result of N fertilization in saline soils, increases the equilibrium concentration of P in soil solution resulting in a larger quantity of mobile P. Consequently, the likelihood of P leaching is enhanced as the soil pH decreases. However, there were no significant differences in water-extractable soil P between N-fertilized and N-unfertilized corn treatments at any depth in the soil (Fig. 1). A second possibility is that since fertilizer-N additions increase above- and belowground biomass production in corn systems (Fehrenbacher and Snider, 1954), P-mineralization processes may release more mobile P from coarse root decomposition under N-fertilized than N-unfertilized agricultural or prairie soils. Additionally, organic matter lining the inside of root channels blocks P-adsorption sites (Ojala et al., 1983) resulting in less attenuation of mobile P as soil leachate solution is redistributed in the profile. Therefore, we speculate that with greater root growth, presumably more macropore flow, and more mobile P released by mineralization under N-fertilized than N-unfertilized soils, leachate solutions may contain higher P concentrations under N-fertilized than N-unfertilized agricultural or prairie soils.
Phosphorus concentrations in leachate generally decreased after crop planting and N fertilization (i.e., early May) as the major drainage–flux period slowed down. From this point on, water loss from the soil, either by soil evaporation plus plant transpiration or drainage, were roughly in equilibrium with precipitation inputs until about the end of July, when transpiration began to exceed precipitation (Brye, 1999; Brye et al., 2000). From May through July, one would expect soil processes to be reasonably stable and that leachate-P concentrations would be related to a source of mobile P at the soil depth sampled by the lysimeters (i.e., 90–120 cm). During this time (i.e., May through July 2000), leachate-MRP concentrations were highly correlated (r = 0.81) to the amount of water-extractable soil P, while inversely correlated to Bray-extractable soil P (r = -0.24), in the 90- to 120-cm depth increment (Table 8). This result suggests that water-extractable soil P may be a better indicator of P leaching potential than Bray-extractable soil P.
|
|
Though the movement of P has really only gained attention in recent years, several previous studies have reported P concentrations in soil leachate solutions collected from a variety of sources, generally from tile drains and occasionally from zero-tension lysimeters, and under a variety of cropping systems (Table 1). Many of these studies have reported dissolved-P concentrations in excess of those recognized as critical threshold concentrations for surface waters, such as approximately 0.01 mg L-1 for lakes, above which noxious aquatic growth may occur (Sawyer, 1947; Wood, 1998). This is of potential concern since soil leachate solution can be hydrologically connected to ground water and surface waters via subsurface flow patterns; thus dissolved-P concentrations in excess of critical limits represent potentially environmentally significant concentrations. However, much attenuation of dissolved P can occur from the time leachate leaves the root zone until the time it reaches ground water. In this study, concentrations of 0.01 to 0.02 and 0.02 to 0.09 mg MRP L-1 were found to leach below the root zones of the natural restored prairie and N-unfertilized corn agroecosystems and the N-fertilized corn agroecosystems, respectively (Tables 5–7).
Overall, mobile P concentrations fall within the range previously reported for agricultural soil leachate solutions, which vary from below detection limits to as high as 1.5 mg MRP L-1 (Table 1). Additionally, the range of leachate-P concentrations found in this study was similar to the range of runoff-P concentrations found from NT and CP corn systems at Arlington (i.e., 0.01 to 0.09 mg MRP L-1), which could be additional evidence suggesting that greater macropore flow results in higher leachate-P concentrations under corn than prairie (Bundy et al., 2001). More importantly, the range of leachate-P concentrations found in this study was higher than P concentrations found in the ground water of the Arlington area (i.e., 0.015 and 0.016 mg L-1 of MRP and TP, respectively; unpublished data). This result suggests that, even though some attenuation of dissolved P occurs after leachate leaves the root zone, background leachate-P concentrations from natural ecosystems can still potentially contribute environmentally significant concentrations of agriculturally related ground water and surface water pollutants 20 to 24 yr after being restored from a long history of cultivated agriculture.
Mobile Phosphorus Leaching
Cumulative mean MRP leaching losses during March through April tended to be higher, though not significantly, for the CPf than the NTf corn agroecosystem due to higher mean drainage, except in 1998, though volume-weighted MRP concentrations tended to be lower, though not significantly, for the CPf than the NTf corn treatments (Table 5). More rainfall reaches the soil surface in the CP than NT treatment presumably because some rainfall is intercepted by the residue remaining on the surface of the NT treatment. Consequently, MRP leaching losses were due to the differences in drainage between the two N-fertilized tillage treatments. Prairie MRP leaching losses tended to be smaller than leaching losses from the corn agroecosystems. Higher MRP loads were generally influenced more by differences in drainage than by differences in MRP concentrations in the prairie and CPf and NTf corn treatments.
Cumulative mean MRP leaching losses for the 21-mo continuous monitoring period were 42, 426, and 467 g MRP ha-1 for the prairie, CPf, and NTf corn agroecosystems, respectively (Fig. 3) . Cumulative mean MRP leaching losses from the N-fertilized corn systems were significantly higher (p = 0.03) than MRP losses from the prairie (Table 6).
|
|
Molybdate-reactive P leaching losses past the 1.4-m soil depth plane were significantly correlated (p < 0.05) to leachate-MRP concentration and OM in the 5- to 15- and 15- to 30-cm depth increments (Table 8). This might indicate that these soil layers are contributing sources to the dissolved P found in the lysimeters at 1.4 m due to P release as a result of OM mineralization. Conversely, it might indicate that P released in the 0- to 5-cm depth is transmitted relatively quickly through the 5- to 30-cm depth due to (i) the presence of OM to block adsorption sites and (ii) greater potential of the dissolved P to encounter well-established macropores that exist below the current or previous plow layer. In the soil depth increment closest to the ETLs, (i.e., 90–120 cm), water-extractable soil P was positively, while Bray-extractable soil P was inversely, though not significantly, correlated to MRP load in leachate. This relationship was also observed for leachate-MRP concentrations, suggesting that a measure of water-extractable soil P better reflects the P concentration and load in leachate solutions. A measure of Bray-extractable soil P at depth, as a possible mobile P source, would have provided an erroneous interpretation of the potential for P leaching. In contrast, both water- and Bray-extractable soil P in the 0- to 5-, 5- to 15-, and 15- to 30-cm depth increments were positively, though not significantly, correlated to leachate-MRP concentration. Nonetheless, it appears that neither a measure of water- nor Bray-extractable soil P in the surface or at depth in the soil could be used to predict P concentrations or loads in leachate.
Soluble P losses of <1 kg P ha-1 yr-1 are common from agricultural mineral soils (Sims et al., 1998). Similar to leachate-MRP concentrations, MRP and TDP leaching losses measured in this study fell within the range of P losses previously reported for agricultural soil leachate solutions (Table 1). Mobile P leaching losses from a restored tallgrass prairie and NT and CP corn agroecosystems without annual inorganic N applications were below 400 g P ha-1 yr-1, suggested by Stamm et al. (1997) as an acceptable critical limit. Conversely, P leaching losses from corn agroecosystems with annual inorganic N additions approached acceptable critical limits for annual P loss. However, acceptable critical limits for P load as identified by Stamm et al. (1997) are site- and water body–specific and are not necessarily related to potentially environmentally significant concentrations of dissolved P since they are in terms of mass per area rather than on a mass-per-volume basis. Nonetheless, P leaching appears to be affected by fertilizer-N additions. Further research will be needed to determine more conclusively the effects of fertilizer-N on P leaching.
|
SUMMARY AND CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONSUMMARY AND CONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. H. Fisher and R. W. Healy Water Movement within the Unsaturated Zone in Four Agricultural Areas of the United States J. Environ. Qual., May 1, 2008; 37(3): 1051 - 1063. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. E. Schilling and J. Spooner Effects of Watershed-Scale Land Use Change on Stream Nitrate Concentrations J. Environ. Qual., October 27, 2006; 35(6): 2132 - 2145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Toth, Z. Dou, J. D. Ferguson, D. T. Galligan, and C. F. Ramberg Jr. Nitrogen- vs. Phosphorus-based Dairy Manure Applications to Field Crops: Nitrate and Phosphorus Leaching and Soil Phosphorus Accumulation J. Environ. Qual., October 27, 2006; 35(6): 2302 - 2312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. O. Nelson, J. E. Parsons, and R. L. Mikkelsen Field-Scale Evaluation of Phosphorus Leaching in Acid Sandy Soils Receiving Swine Waste J. Environ. Qual., October 12, 2005; 34(6): 2024 - 2035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Erickson, J. L. Cisar, G. H. Snyder, and J. C. Volin Phosphorus and Potassium Leaching under Contrasting Residential Landscape Models Established on a Sandy Soil Crop Sci., January 31, 2005; 45(2): 546 - 552. [Abstract] [Full Text] [PDF]
|
|
