| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a USDA-ARS, Pasture Systems and Watershed Management Research Unit, 3702 Curtin Road, University Park, PA 16802-3702
b Department of Soil Science, North Carolina State University, Raleigh, NC 27695
* Corresponding author (peter.kleinman{at}ars.usda.gov).
Received for publication March 21, 2003.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: DRP, dissolved reactive phosphorus • EDI, effective depth of interaction • SS, suspended solids • TP, total phosphorus • WEP, water-extractable phosphorus
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The processes by which agricultural soils, and, more specifically, manure management, influence the transport of P in agricultural runoff are well documented. Over the long term, application of manure P to soils at rates greater than annual crop removal results in the accumulation of P in surface soil (Smith et al., 1998). Elevated concentrations of soil P affect water quality through desorption of soluble P forms to runoff water (Sharpley et al., 1981a) as well as through erosion, which preferentially removes soil particles that are enriched in P relative to bulk soil (Sharpley, 1985b). Over the short term (months), application of manure to soils, particularly via broadcasting, temporarily elevates P available to runoff water, due to high concentrations of water-soluble P in manure (Edwards and Daniel, 1993). In addition, low-density organic matter fractions in manure (flocs) are highly susceptible to erosion when manure is broadcast (McDowell and Sharpley, 2002).
The study of P transport in surface runoff from agricultural soils has relied on a variety of research methods. Watershed monitoring represents the most direct evaluation of soil and management effects on water quality because watershed export of P is ultimately the concern to eutrophication. However, only a limited number of studies, mostly of smaller watersheds, have convincingly linked soil and manure management to watershed P export (Sharpley et al., 1991; Smith et al., 1991). Interactions of hydrology (surface and subsurface flow), climate, geomorphology, soils, and management tend to mask causal links between field management and watershed P export (Calhoun et al., 2002).
Field runoff plots of various sizes (2–622 m2) have been used effectively, in conjunction with either natural or simulated rainfall, to relate soil and manure management to runoff water quality (McDowell and Sharpley, 2001; Gascho et al., 1998; Zhao et al., 2001). Field runoff plots provide control of many landscape variables that potentially confound watershed research. In addition, large numbers of replicated treatments are possible with field plots, facilitating quantitative evaluation and comparison of alternative treatments. For instance, to develop defensible environmental thresholds for P levels in agricultural soils, researchers from at least 29 states are participating in the National Phosphorus Research Project (NPRP), using rain simulators, 2-m-long runoff plots, and a common experimental protocol to quantify soil-specific relationships between soil P and P in runoff (Sharpley et al., 1999, 2002b).
Runoff boxes, typically packed with soil and subjected to simulated rainfall (subsequently referred to as packed boxes), allow for even greater control of confounding variables than do field runoff plots, as soils can be homogenized to minimize significant variability in physical and chemical characteristics. As with field plots, packed boxes have been used to quantify soil P–runoff P relationships (Sharpley, 1995; Pote et al., 1999). However, packed boxes are least representative of field and landscape conditions. The hydrology of sieved, packed soil boxes is undoubtedly different from field soils with intact structure, complex horizonation, and the complete array of fine-earth and coarse fragments. In addition, recent studies of P transport using packed boxes have generally relied on bare soils that are highly susceptible to erosion (Sharpley, 1995), in contrast with field plot studies that have included a variety of soil cover and cultivation treatments (Edwards and Daniel, 1993; Torbert et al., 2002; Zhao et al., 2001).
Recently, a series of findings, primarily from packed box studies, have provided the quantitative basis for developing P availability coefficients in some P site assessment indices (Sharpley et al., 2003). Phosphorus availability coefficients are quantitative indicators of the relative availability of P in mineral fertilizer or manure to be transported in runoff (Leytem et al., 2003). Kleinman et al. (2002a) observed that concentrations of DRP in runoff from packed boxes recently broadcast with various manures or mineral fertilizer at the same rate of TP addition were a function of the WEP concentration of the applied P source. These findings were extended by Kleinman and Sharpley (2003), who examined application rate and timing effects related to WEP in manure, as well as by Brandt and Elliott (2003), who examined runoff P losses from soils that were broadcast with various biosolids. All studies used shallow (5-cm-deep) packed boxes with infiltration properties possibly controlled by box depth rather than soil properties, even though the packed boxes did allow for some drainage via nine, 5-mm-diameter drain holes (Kleinman et al., 2002a).
Use of shallow soil boxes with limited infiltration may affect conclusions regarding manurial P transport. Sharpley (1985a), in experiments using packed soil boxes, reported effective depths of interaction (EDI) between runoff water and soil from 0.1 to 3.7 cm, highlighting the importance of processes affecting P distribution at the soil surface. Elsewhere, Pote et al. (2001) observed that DRP concentration in runoff from field plots broadcast with swine slurry was negatively correlated with infiltration rate. They hypothesized that increasing infiltration resulted in greater translocation of soluble P from the manure below the EDI, into the soil subsurface, where it was unavailable to runoff. Thus, it is possible that poor infiltration resulting from shallow-bottomed soil boxes with restricted water holding capacity could limit translocation of manure P into the soil, resulting in P transport that does not adequately reflect natural soil controls.
Given that results from grassed field plots and bare soil boxes are used interchangeably to calibrate P site assessment indices, the objective of this study was to examine the use of packed boxes in the study of P transport from agricultural soils. Specifically, this study was conducted to (i) compare results from unamended grassed field plots with boxes packed with bare soil, particularly with regard to the relationship between DRP in runoff and soil P, and (ii) determine if conclusions regarding manure management effects on runoff P concentrations derived from packed box experiments are influenced by box depth.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Comparison of Field Plots and Packed Boxes: Experiment 1
Field Plots
Field runoff experiments were conducted on plots established in Hartleton and Honeoye soils representing a broad range of Mehlich-3 P concentrations (44–386 mg kg–1 for Hartleton soils, 13–136 mg kg–1 for Honeoye soils). Slope gradients varied from 3 to 8%. All soils had established stands of mixed grasses (orchard grass, Dactylis glomerata L.; timothy, Phleum pretense L.; tall fescue, Festuca arundinacea Schreb.; white clover, Trifolium repens L.) or alfalfa (Medicago sativa L.), cut to a 7-cm height, and had not received manure or mineral fertilizer in the six months before the runoff experiment. A total of 8 plots were established on Hartleton soils and 18 plots on Honeoye soils. At each location, one pair of 2-m-long and 1-m-wide runoff plots was installed, isolated on the upper three sides by steel frames driven 5 cm into the soil and extending 5 cm above the soil. At the lower end of each plot, a gutter was installed, inserted 5 cm into the soil with the upper edge level with the soil surface. The gutter was equipped with a canopy to exclude direct input of rainfall and a 2-cm plastic tube was used to route runoff water from the gutter to plastic collecting vessels.
Rain simulations were conducted on two successive days following the protocol of the National Phosphorus Research Project (2001). Portable rain simulators (Humphry et al., 2002) equipped with TeeJet 1/2 HH SS 50 WSQ nozzles (Spraying Systems Co., Wheaton, IL) were placed approximately 3 m above the soil surface. At this height, simulated rainfall achieves approximately 90% terminal velocity and has a coefficient of uniformity of >0.80 within the 2- x 2-m area directly below the nozzle. On each day, rainfall was delivered at approximately 75 mm h–1 until 30 min of runoff was collected. Following each simulation, runoff water was thoroughly stirred to resuspend settled particles and immediately sampled. A filtered (0.45 µm) subsample was obtained within 24 h. Runoff samples were stored at 4°C before laboratory analysis.
While antecedent soil moisture was expected to range widely between plots before the first event, soils were expected to be at field capacity before the second event, as confirmed by capacitance sensor (Theta Probe; Delta-T Devices, Cambridge, UK). Because variability in antecedent soil moisture affects both hydrology and P transport, results from only the second runoff event were used to assess trends in P transport related to soil P.
Packed Runoff Boxes
Rainfall simulations were conducted following the National Phosphorus Research Project packed-box protocol (National Phosphorus Research Project, 2001). This protocol uses 1-m-long x 20-cm-wide x 5-cm-deep stainless steel boxes, with back walls 2.5 cm higher than the soil surface, and 5-mm diameter drainage holes in the base. Cheesecloth was placed on the bottom of each box before soils were packed. At the lower end of each box, a gutter equipped with a canopy channeled runoff water to collection containers (Kleinman et al., 2002a).
For this experiment, surface horizons (0–20 cm) of Hartleton and Honeoye soils representing a variety of soil test P concentrations were collected, field-sieved to pass through a 1.4-cm-diameter opening, air-dried, and thoroughly mixed. To ensure homogeneity of the individual soils, the effectiveness of mixing was evaluated by conducting Mehlich-3 P extraction on six subsamples from each soil and determining the coefficient of variation (standard deviation divided by mean Mehlich-3 P concentration) for each soil. For both soils, the coefficient of variation was <0.05. Soils were packed into boxes to achieve an approximate bulk density of 1.3 to 1.5 g cm–3.
Packed boxes (N = 8 for each soil) were placed under the rain simulator, inclined to a 3% slope gradient, and staggered so that, during rainfall simulation, splash from one box would not be intercepted by another box. Soils were first saturated using the rainfall simulator (75 mm h–1 until ponding was observed, approximately 10 min) and allowed to drain for 72 h before the initial rainfall event. All soils were approximately at field capacity at the start of the first runoff-generating event, ensuring that hydrologic variability related to antecedent moisture was minimized. Rain simulations and runoff collection procedures followed those described for the field plots.
Effect of Box Depth and Manure Application on Phosphorus Transport: Experiment 2
Interactions among box depth, broadcast manure, and timing and sequence of runoff event on runoff P losses were assessed using a modified version of the packed box protocol described above. For this experiment, an additional set of boxes was constructed, with all features similar to the National Phosphorus Research Project boxes except that the modified boxes were 25 cm deep.
Surface horizons of low-P Hartleton (average Mehlich-3 P = 16 mg kg–1) and Honeoye soils (average Mehlich-3 P = 21 mg kg–1) were collected, processed in the fashion described above, and analyzed for Mehlich-3 P. Following mixing, the coefficient of variation for Mehlich-3 P of six randomly selected samples was <0.05 for both soils. Soils were packed into the boxes to obtain a bulk density of 1.3 to 1.5 g cm–3. For each soil, eight 5-cm-deep packed boxes and eight 25-cm-deep packed boxes were used.
Three manures were selected to represent a range of animal species, dry matter contents, and P solubilities. Dairy manure, layer poultry manure, and swine slurry were collected, thoroughly mixed, and stored at 4°C for a maximum of one week before analysis. Dairy manure and swine slurry were sampled from the Pennsylvania State University Dairy and Swine Centers at University Park, PA. The dairy manure was from lactating Friesian-style dairy cows (Bos taurus) and was scraped from a free stall barn. Swine slurry was from finishing sows (Sus scrofa domestica) that was washed into a holding tank and agitated before sampling. Poultry (Gallus gallus domestica) manure was from a laying operation in Northumberland County, PA, and was collected directly from the layer house.
Rainfall-runoff simulations were performed before and after manure was broadcast onto the packed boxes following the basic rain simulation and runoff collection protocol described earlier for packed soil boxes (75 mm h–1 rainfall, 30-min runoff). Before manure application, two rainfall simulations were conducted on consecutive days to assess trends in runoff P derived from bare soil P only. Three days after the second event, dairy manure, poultry manure, and swine slurry were broadcast onto individual packed boxes at a rate corresponding to 100 kg TP ha–1. A control treatment (zero manure application) was left for comparison. Each treatment was conducted in duplicate. Consecutive rainfall-runoff simulations were conducted three and four days after the manure was applied. Runoff was collected, processed, and analyzed as described above.
To assess possible differences in soil moisture related to box depth, volumetric soil water content was measured by capacitance sensor. Before and after each rainfall simulation event, two measurements were obtained from the top and bottom ends of every packed box (0- to 4-cm depth), with special attention paid to minimizing disturbance during insertion of the capacitance sensor.
After the last rainfall simulation event, soils from each box were sampled to assess soil P accumulation with depth. For the 5-cm-deep boxes, 0- to 1-, 1- to 3-, and 3- to 5-cm depth increments were sampled. For the 25-cm-deep boxes, additional depth increments of 5- to 10- and 10- to 25-cm were sampled.
Chemical Analyses
Soil Analysis
Soils used in packed box experiments were sampled before the rainfall simulations for Mehlich-3 P analysis. In addition, for each field plot, ten 5-cm-deep soil samples were collected with a 2-cm-diameter stainless steel probe following the rainfall simulations and mixed thoroughly to provide a composite soil sample. All soils were air-dried, sieved (2 mm), and analyzed for Mehlich-3 P by shaking 2.5 g of soil with 25 mL of Mehlich-3 solution (0.2 M CH3COOH + 0.25 M NH4NO3 + 0.015 M NH4F + 0.013 M HNO3 + 0.001 M EDTA) for 5 min (Mehlich, 1984). Extract P was determined colorimetrically, by a modified method of Murphy and Riley (1962), with a spectrophotometer wavelength of 712 nm. Soil pH was determined by mixing air-dry soil with distilled water (5 g to 5 mL).
Soil samples collected from packed boxes at the conclusion of Experiment 2 were air-dried, sieved (2 mm), and analyzed for Mehlich-3 P and WEP. Water-extractable soil P was measured by shaking 0.5 g of soil in 5 mL of distilled water for 1 h, filtering the supernatant through a Whatman (Maidstone, UK) no. 1 paper filter, and determining P colorimetrically.
Runoff Water Analysis
Dissolved reactive P was determined on 0.45-µm-filtered runoff water by the colorimetric method described for soil extracts. Total P was measured on unfiltered runoff water by modified semimicro Kjeldahl procedure of Bremner (1996). Runoff water was also analyzed for suspended solids (SS) by evaporating 200 mL of unfiltered runoff water in an oven at 70°C and weighing the remaining material.
Manure Analysis
Manure was analyzed for TP by modified semimicro Kjeldahl procedure (Bremner, 1996). Water-extractable P was analyzed by the method of Kleinman et al. (2002b). One gram dry-weight equivalent fresh manure was shaken with 200 mL of distilled water on an end-over-end shaker for 60 min. The mixture was then centrifuged (about 2900 x g for 20 min to facilitate filtration) and filtered through a Whatman no. 1 filter paper. Filtrate P was determined colorimetrically. Manure pH was measured after mixing 1 g (equivalent dry weight) of fresh manure with 100 mL distilled water. Dry matter content of all manures was determined gravimetrically after oven-drying manures at 70°C for 48 h.
Statistical Analyses
Runoff P concentrations (mg L–1) and losses (kg ha–1) were logarithmically transformed to conform with assumptions of normality and equal error variances. As DRP and TP were often less than one (mg L–1 or kg ha–1), these variables were transformed for analysis by adding 1 to the P concentration and determining the logarithm of that sum so that no negative values were obtained (Neter et al., 1996). Treatment effects were evaluated by one-way ANOVA for the field plot–packed box comparisons (Experiment 1) and by general linear model for the box–depth comparisons (Experiment 2), along with Tukey's mean separation. Relationships between soil and runoff variables were quantified by least squares regression, and differences in regression parameters were assessed by a homogeneity of variance test (Gomez and Gomez, 1976). Treatment differences discussed in the text were significant at 
0.05. Analyses, with the exception of homogeneity of variance of regression coefficients, were performed using Minitab's statistical software (Release 13; Minitab, 2001) and SAS Version 8 (SAS Institute, 1999).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Infiltration was significantly greater in the field plots than in the packed boxes (Table 1), but did not differ significantly between soils, which had similar particle size distributions. Differences in infiltration between field plots and packed boxes probably reflect the role of preserved soil structural attributes, such as intact macropores, that were not present in the sieved soils of the packed boxes (Quisenberry and Phillips, 1976). Surface sealing due to aggregate dispersion by direct raindrop impact also probably reduced infiltration into the bare soils of the packed boxes (McIntyre, 1958). Results from Experiment 2, described below, suggest that the drainage design of the packed boxes (nine 5-mm drainage holes) did not significantly impede infiltration into the sieved soils. Indeed, rainfall infiltration into packed soil boxes persisted throughout the runoff event, as runoff depths (25.0–36.5 mm) did not achieve 100% of rainfall (37.5 mm) over the 30-min runoff event (Table 1).
Differences in infiltration clearly affected runoff depth, which was negatively related to infiltration for both field plots (runoff = 3.3 – 0.4 x infiltration; r2 = 0.68) and packed boxes (runoff = 3.6 – 0.5 x infiltration; r2 = 0.76). Significantly less runoff was produced from the field plots than from the packed boxes, and no significant differences were observed between soils (Table 1).
Suspended Solids, Total Phosphorus, and Dissolved Reactive Phosphorus Concentrations in Runoff
As expected, runoff from field plots and packed boxes contrasted with regard to SS concentration (g L–1), which was greater from the packed boxes than from the field plots (Table 1). These differences reflect the presence of a protective grass or alfalfa canopy in the field plots, compared with the exposed, bare soil of the packed boxes. In addition, sieving and packing soils into boxes destroys larger soil aggregates, increasing the availability of fine particles to runoff, and possibly decreasing the stability of remaining aggregates. Significant differences in SS concentration were also detected between the two soils, with greater SS concentrations from the Hartleton soil than from the Honeoye soil for both field plots and packed boxes.
Runoff TP concentrations were strongly related to SS concentrations in runoff from Hartleton field plots and weakly related to SS concentrations from Honeoye soil boxes (Table 2), reflecting the importance of particulate P to TP concentrations in runoff. Even though particulate P was not directly measured in this study, particulate P probably accounted for most of the difference between TP and DRP in runoff. For all packed boxes, DRP contributed from 1 to 8% of TP in runoff, whereas for all field plots, DRP contributed 5 to 38% of TP. The larger contribution of DRP to TP in runoff from field plots reflects the lower erosion from the grass-covered field plots than from the bare soils of the packed boxes and possibly dissolved P release from plant residue at the surface of the field plots.
|
|
Comparison of runoff DRP–soil P trends between soils, as well as between field plots and packed boxes, can be biased by unequal ranges of Mehlich-3 P concentrations. In Experiment 1, the ranges of Mehlich-3 P concentrations of the Honeoye soil were considerably narrower (13–136 mg kg–1 for the field plots; 21–80 mg kg–1 for the packed boxes) than those of the Hartleton soil (44–386 mg kg–1 for the field plots; 16–410 mg kg–1 for the packed boxes). McDowell and Sharpley (2001) identified nonlinear relationships between DRP concentrations in runoff and Mehlich-3 P, with a Mehlich-3 P threshold of approximately 200 mg kg–1 separating linear regressions of different slopes. Their results suggest that a range of Mehlich-3 P concentrations falling on one side or the other of the threshold would skew linear regression, such that soils with Mehlich-3 P below the threshold would produce a significantly lower regression slope than soils above the threshold. However, in this study, no consistent differences in regression slopes were observed on that basis and regression slopes for Honeoye soils were not significantly different from those obtained from Hartleton soils (Table 2). Indeed, other runoff studies that have included broad ranges of Mehlich-3 P concentrations in a variety of soils (Sharpley, 1995; Torbert et al., 2002) have similarly reported linear relationships between runoff DRP and Mehlich-3 P, indicating that nonlinear trends are not universal.
Normalizing Runoff Properties to Address Variability in Runoff, Contributing Area, and Rainfall
Because variability in runoff, rainfall, and contributing areas is common to many studies of DRP transport, we compared the effects of different normalization approaches on relationships between DRP (mg) and Mehlich-3 soil P in runoff for all data obtained from Experiment 1. Specifically, DRP mass in runoff was divided by catchment area, runoff, rainfall, area x runoff, area x rainfall, runoff x rainfall, and area x runoff x rainfall. As summarized in Table 3, normalizing procedures resulted in widely differing regression equations and r2 values. Normalizing by catchment area provided the best r2 values for regressions, illustrating the importance of this variable to DRP loading in runoff. Normalizing on the basis of rainfall depth consistently resulted in the lowest r2 values. As rain simulation events were controlled for runoff duration (30 min), and runoff from plots and boxes tended to reach equilibrium flow and DRP concentration within 15 min of runoff initiation (Sharpley et al., 1981a; Sharpley and Kleinman, 2003), variability in rainfall depth was not expected to play a dominant role in P release from unamended soil.
|
Another common means of presenting runoff data is as mass exported per standardized contributing area, referred to as "loss" (kg ha–1). When runoff results from Experiment 1 were calculated in this way, conclusions regarding field plot and packed box trends were consistent with those derived from concentration data (g kg–1 or mg kg–1). Specifically, for a given soil, SS and TP losses were greater from packed boxes than from field plots and DRP losses were similar between field plots and packed boxes (Table 1). In addition, DRP losses in runoff were strongly related to Mehlich-3 soil P (Fig. 1c). As with DRP concentration, a single regression equation predicted DRP losses when all data were combined [log(DRP loss + 1) = 0.002 + 0.0001 x Mehlich-3 P; r2 = 0.82). Thus, differences in runoff depths and catchment areas of field plots and packed boxes, as they affected losses of SS, TP, and DRP, did not significantly alter conclusions drawn from concentration data.
Experiment 2: Effect of Box Depth and Manure Application on Phosphorus Transport
Influence of Box Depth on Phosphorus Transport from Bare Soils
Properties of both soils used in the packed box depth experiments were similar, with the exception of pH, which was expected due to differing mineralogies (Table 4). Rainfall depths varied between soils and box depths (Table 5), with the most rainfall applied to the Honeoye soils on the first runoff event and generally more rainfall applied to the 25-cm boxes than to the 5-cm boxes. Hydrology of the bare soils did not differ consistently between box depths. Infiltration, a key concern given the possibility that the 5-cm-deep boxes create an artificially perched water table, was not significantly different across events or between box depths for the Hartleton soil. While 25-cm-deep boxes allowed greater infiltration than 5-cm-deep boxes for the Honeoye soil, differences were significant only for the first event. Differences in runoff depths were also inconsistent between soils (Table 5), although more runoff was generally produced during the second event than the first event due to greater soil moisture at the start of the second event (data not shown). Indeed, soil moisture was one variable that behaved consistently across box depth treatments; no significant differences in moisture of the upper 4 cm of soil were observed between 5- and 25-cm boxes, either before or after any of the rainfall events (data not shown).
|
|
Influence of Box Depth on Phosphorus Transport from Soils Broadcast with Manure
Properties of the three manures ranged widely (Table 6). Total nitrogen (TN) to TP ratios were 6.2:1, 2.4:1, and 4.5:1 for the dairy manure, poultry manure, and swine slurry, respectively. Thus, an N-based manure application rate for silage corn of 300 kg TN ha–1 (Beegle, 1999) would result in TP application rates of 48, 125, and 67 kg ha–1 for the dairy manure, poultry manure, and swine slurry, respectively. Water-extractable P (dry weight equivalent) was most concentrated in the swine slurry, with concentrations in dairy and poultry manures similar. As a percentage of TP concentration, WEP was roughly 60% of dairy manure TP, 18% of poultry manure TP, and 71% of swine slurry TP.
|
|
As with bare soils before manure application, box depth did not significantly affect DRP or TP concentrations in runoff after manure application (Table 7). As DRP concentrations were largely a function of WEP in applied manures, one possibility was that differential translocation of soluble P from the manure into the soil would result in different concentrations of P in the layer of manure and soil interacting with runoff water (the EDI). This was particularly of concern for the manures with high water content, such as the swine slurry, which contained only 3% solids, as immediate infiltration of water from manure could account for substantial translocation of soluble manure P out of the EDI. According to this hypothesis, shallow boxes would prevent soluble P from fully infiltrating into the soil, resulting in artificially elevated concentrations of soluble P at the soil surface that would be prone to runoff. The effect would be exaggerated by differences in rainfall depths between treatments (Table 7), with lower DRP concentrations expected from treatments subjected to greater rainfall. Sharpley (1985a) concluded that soil slope, rainfall intensity, and erosion were the dominant controls of EDI in unamended soils. In this study, slope (3%) and rainfall intensity (75 mm h–1) were held constant and erosion did not differ significantly between box depths. In addition, there were few significant differences in infiltration and runoff between treatments in the final two events, and observed differences were inconsistent (Table 7).
Examination of P distribution in soils after the fourth rainfall event showed few differences between 5- and 25-cm-deep boxes, suggesting that the fate of applied P was not affected by box depth. High concentrations of P were clearly translocated from the broadcast manures into the upper 1 cm of soil, as evidenced by the elevated WEP and Mehlich-3 P of manured soils compared with unmanured soils and subsoils (Table 8). Statistically significant increases in these properties were observed in the 0- to 1-cm soil samples of the dairy manure and swine slurry treatments only. Although WEP and Mehlich-3 P were also somewhat elevated in the 0- to 1-cm soil samples of the poultry manure treatment, they were not significantly different from the subsoil. No significant differences in WEP and Mehlich-3 P were evident at lower depths indicating that translocation of manure P was primarily restricted to the upper 1 cm of soil (Table 8).
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This study also examined whether limited infiltration in shallow soil boxes, as compared with deeper boxes, substantially affects conclusions related to P transport from manured soils. A key concern was that perched water tables in shallow boxes would impede translocation of soluble P from broadcast manures, inflating soluble P release to runoff water and biasing conclusions regarding the control of manure WEP on runoff P. However, depth of packed soil boxes did not affect P concentrations in runoff. The application of high concentrations of WEP in manure to the surface of packed boxes resulted in similar increases in runoff P from 5- and 25-cm-deep boxes. Although significant concentrations of P were translocated from the dairy manure and swine slurry into the surface soil, trends were similar between boxes of different depths. Furthermore, no significant differences were observed below a 1-cm depth, indicating that, at least in packed soil boxes, translocation of manure P below the EDI is not significant.
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  M. S. Srinivasan, P. J. A. Kleinman, A. N. Sharpley, T. Buob, and W. J. Gburek Hydrology of Small Field Plots Used to Study Phosphorus Runoff under Simulated Rainfall J. Environ. Qual., October 24, 2007; 36(6): 1833 - 1842. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Little, S. C. Nolan, J. P. Casson, and B. M. Olson Relationships between Soil and Runoff Phosphorus in Small Alberta Watersheds J. Environ. Qual., July 17, 2007; 36(5): 1289 - 1300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. K. Udeigwe, J. J. Wang, and H. Zhang Predicting Runoff of Suspended Solids and Particulate Phosphorus for Selected Louisiana Soils Using Simple Soil Tests J. Environ. Qual., July 17, 2007; 36(5): 1310 - 1317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Kleinman, D. Sullivan, A. Wolf, R. Brandt, Z. Dou, H. Elliott, J. Kovar, A. Leytem, R. Maguire, P. Moore, et al. Selection of a Water-Extractable Phosphorus Test for Manures and Biosolids as an Indicator of Runoff Loss Potential J. Environ. Qual., July 17, 2007; 36(5): 1357 - 1367. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Franklin, C. Truman, T. Potter, D. Bosch, T. Strickland, and C. Bednarz Nitrogen and Phosphorus Runoff Losses from Variable and Constant Intensity Rainfall Simulations on Loamy Sand under Conventional and Strip Tillage Systems J. Environ. Qual., May 7, 2007; 36(3): 846 - 854. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Elliott, R. C. Brandt, P. J. A. Kleinman, A. N. Sharpley, and D. B. Beegle Estimating Source Coefficients for Phosphorus Site Indices J. Environ. Qual., October 27, 2006; 35(6): 2195 - 2201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
