| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
USDA-ARS, Water Conservation Laboratory, 4331 E. Broadway Rd., Phoenix, AZ 85040
* Corresponding author (Hadi.Tabbara{at}asu.edu)
Received for publication February 18, 2002.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: BAP, bioavailable phosphorus • DOP, dissolved organic phosphorus • DRP, dissolved reactive phosphorus • PP, particulate phosphorus • TDP, total dissolved phosphorus • TP, total phosphorus • TSS, total suspended solids
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Manure contains both organic and inorganic P compounds of varying solubility. The proportions are dependent on, among other factors, manure type and animal diet (Barnett, 1994). Phosphorus is also partitioned into the solid and liquid phases of manure with more than 90% of TP in swine manure associated with solids (Gerritse, 1977) mostly in the inorganic form (Fordham and Schertmann, 1977; Van Riemsdijk et al., 1984). Gerritse (1981) reported that in swine slurry, inorganic P in solution accounted for 5% (100 mg L-1) and organic P in solution for 1% (20 mg L-1) of TP. More recently, Sharpley and Moyer (2000) found that 90% of TP in swine slurry is in the inorganic form. They reported that the water-extractable form constituted 18% of TP in the inorganic and 5% in the organic fractions, while a large proportion (55%) of inorganic P was hydroxide soluble.
Phosphorus losses from soils are influenced primarily by P source and chemical form (Dunigan and Dick, 1980; Edwards and Daniel, 1993; Eghball and Gilley, 1999; McLeod and Hegg, 1984; Nichols et al., 1994; Sharpley and Sisak, 1997; Withers et al., 2001), tillage and P placement (Baker and Laflen, 1982; Bundy et al., 2001; Ginting et al., 1998; Kimmell et al., 2001; Mostaghimi et al., 1992; Mueller et al., 1984; Sharpley et al., 1993), rate and timing of application (Edwards and Daniel, 1993; Khaleel et al., 1980; Smith et al., 1998), intensity and timing of rainfall events (Edwards and Daniel, 1993; Mueller et al., 1984), and soil P level (Pote et al., 1996; Sharpley, 1995). However, P losses to runoff water depend less on soil P concentrations and more on application rate and method and rainfall intensity if rainfall occurs soon after manure application (Edwards and Daniel, 1993; Sauer et al., 2000; Sharpley and Tunney, 2000). A recent report highlighted the risk due to rapid incidental losses of P, through surface and subsurface pathways, when manure or fertilizer is surface-applied to grassland soil followed by intermittent rain (Preedy et al., 2001).
Swine manure can be handled as a solid, liquid (less than 4% solids), or slurry (4–10% solids) depending on storage and manure handling system. Liquid and slurry application have the advantage of cost-effectiveness, availability of equipment to handle large volumes, and greater conservation of nutrients (Comfort et al., 1987; Ross et al., 1979). Slurry application is the predominant form for land application in the U.S. Corn Belt, especially in large-scale concentrated feeding operations. It is usually applied in the fall when labor is available and to avoid soil compaction associated with spring application. However, farmers try to avoid fall application to comply with conservation regulations requiring maximum residue cover for spring planting. One alternative best management practice is to apply manure in the summer with injection or incorporation within 24 h after application to avoid intense summer rainstorms. This requires that farmers find a suitable land use for manure in the summer.
Slurry application can also be quite risky since P availability and solubility in runoff is greater compared with solid manure and since the slurry might seal the soil surface, resulting in increased runoff and erosion rates (Haraldsen and Sveistrup, 1996; Ross et al., 1979; Withers et al., 2000). Other studies, however, showed that surface sealing is a short-term effect lasting a few hours and that a longer-term soil protection effect, comprised of physical protection by manure particles from raindrop impact and improved aggregate stability, causes a net increase in hydraulic conductivity and decrease in runoff (Smith et al., 2001). There are also conflicting reports on the effect of tillage methods and soil incorporation of manure on sediment-bound or particulate phosphorus (PP) and TP concentration and losses in runoff water. Some researchers reported that management practices such as tillage and incorporation of manure, by decreasing infiltration and increasing sediment loss from the disturbed soil, result in lower DRP concentrations and higher PP and TP concentrations or loads in runoff water (Bundy et al., 2001; Eghball and Gilley, 1999; Mostaghimi et al., 1992). Others reported that incorporation places P below the thin mixing zone of interaction between the soil and surface runoff and thus decreases PP and TP concentrations and losses in runoff water (Baker and Laflen, 1982; Dunigan and Dick, 1980, Kimmell et al., 2001; Withers et al., 2001).
The objective of this study was to test whether incorporation of liquid swine manure 24 h before an intense rainstorm is effective in reducing soluble as well as sediment-bound P loss in runoff water from sloping land with little or no crop cover. A rainfall simulation study compared TSS, DRP, TP, and bioavailable phosphorus (BAP) concentrations and losses in runoff water after liquid swine manure and liquid inorganic fertilizer were broadcast or incorporated.
The experiment did not include a control; comparisons made in this study are therefore among treated plots and are not relative to a true control.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Liquid swine manure was obtained from the Swine Breeding Research Station near Madrid, Iowa. Manure samples contained 90 g kg-1 total solids (TS), 3450 mg L-1 total Kjeldahl N (TKN), and TP of 1994 mg L-1. Analysis for total Kjeldahl nitrogen (TKN) was determined colorimetrically following automated block digestion as described in USEPA Method 351.2 (USEPA, 1983), and TP of the same digest was determined by Method 2.026 of the AOAC (Association of Official Analytical Chemists, 1980). Dissolved reactive P (Murphy and Riley, 1962) in manure was analyzed after passing an aliquot of the sample through a 0.45-µm membrane filter. Based on targeted rates of 168 and 336 kg N ha-1 (representing the N level normally applied to corn in the U.S. Midwest and twice that level) and plot area of 13.9 m2, the volumes of required liquid manure were 68 and 136 L (5230 and 10460 gallons acre-1) applied to the low- and high-rate plots, respectively. To ensure similar antecedent soil moisture contents in all the plots after the application of liquid manure or liquid fertilizer and before applying rain, water was added to the manure or fertilizer to make the water content equivalent to that supplied at the high manure rate. The rates applied, however, differed from calculated values due to manure heterogeneity in the delivery tank caused by settling, variability in sampling and analysis, and ammonia volatilization. Actual composition and rates of applied manure are summarized in Table 1. The liquid inorganic fertilizer used was ammonium polyphosphate. It consists of a mixture of ortho- and pyrophosphate (two P-atom polyphosphate) and normally is applied preplant to a variety of crops. Total P in ammonium polyphosphate fertilizer was analyzed by USEPA Method 365.4 (USEPA, 1983), DRP by the method of Murphy and Riley (1962), and TKN by the same method used for manure analysis. Quantities of manure or fertilizer plus water were added to a 300-L cone-shaped plastic container raised by a tractor to provide gravity flow and were uniformly hand-applied through a hose by walking once up the slope of each plot. Before application, all plots were tilled parallel to the slope with a tandem disk. The plots receiving the incorporated treatment were disked again to a depth of approximately 0.15 m to incorporate the manure or fertilizer. Thus, this experiment simulates conventional tillage conditions where amendments are either surface-broadcast or incorporated after tillage.
|
Plots were isolated with metal borders driven about 50 mm into the ground, and the joints were sealed with soil berms to prevent leakage. A triangular metal collector 1.5 m wide was installed at the downslope end of each plot, while another collector was laid face-down on top to prevent rainfall from falling on the bottom collector. The collectors delivered runoff to PVC cups connected to PVC pipes that delivered the water over the lower terrace about 2 m from the end of the plots. Six to eight flow rate measurements were made gravimetrically for each plot during runoff by roughly filling a 5-L bucket with runoff water, weighing it, and recording the time interval for filling it. One measurement was made at the beginning of the runoff event with the beginning time recorded, and then measurements were made at 5, 15, 25, 40, 55, and 70 min. Runoff water samples were collected for sediment and chemical analysis by periodically passing the mouth of a 1-L glass mason jar under the runoff outflow during the interval between each two successive flow rate measurements, representing composite grab samples except for the first sample, which was poured directly from the bucket used for the first runoff rate measurement and represented the first 2 to 3 min of the onset of runoff. One 0.25-L subsample from each jar was filtered (0.45 µm) within 2 h of collection, analyzed within 24 h for DRP, and later analyzed for total dissolved phosphorus (TDP) by procedures outlined below. Another 0.25-L subsample (unfiltered) was stored for later analysis for TP and BAP. All unfiltered samples were stored at 4°C and analyzed within 5 wk of collection. Runoff volume and P concentrations were used to determine mass losses of all P forms in runoff from each plot.
Phosphorus in all runoff water samples was analyzed colorimetrically by the Murphy and Riley procedure (1962) with a Quickchem 1000 system (Lachat Instruments, Milwaukee, WI). To measure DRP, samples were centrifuged at 10 000 x g for 20 min and then filtered through a 0.45-µm membrane filter before P analysis. The TDP in filtered samples was determined by persulfate digestion in a block digester (Method 424 C; American Public Health Association, 1985). Determination of TP on unfiltered samples was performed according to USEPA Method 351.2 (USEPA, 1983). Phosphorus was determined from a calibration curve constructed from standards that were digested and treated similarly to the samples. The BAP in unfiltered runoff water was determined by adding from 20 to 35 mL of deionized water to an aliquot containing no more than 0.5 mg P as TP and shaking for 16 h with one iron oxide–impregnated filter. Phosphorus adsorbed onto the filter strips was then extracted with acid (Sharpley, 1993). Standard solutions made up from KH2PO4 were also extracted with one filter paper; recovery was complete within the range of 0.5 to 20 mg P L-1. These strips were used to construct the calibration curves used to determine P in the BAP extract. Dissolved organic phosphorus (DOP) was obtained as the difference between TDP and DRP measured for filtered samples. Particulate phosphorus (PP) was obtained from the mass balance in the equation below. It states that the amount of total P in the collected runoff sample is equal to that of the sum of TDP and the P associated with sediments (PP) in that sample:
 |
To determine TS, a 10-mL aliquot of the unfiltered sample was evaporated in a weighed dish and dried to constant weight in an oven at 103 to 105°C; the increase in weight over that of the empty dish represents the TS. Total dissolved solids (TDS) was determined similarly by drying a 20-mL aliquot of the filtered samples. The difference TS minus TDS represents total suspended solids (TSS), a measure of the sediment content of runoff.
Losses (kg ha-1) of all P forms were computed from the product of the concentration of P in mg L-1 and the depth of flow for each runoff sample (kg ha-1 = mg L-1 x dm) where dm = cm/10. Total loss is the algebraic sum over all samples, while the flow-weighted concentration is obtained by dividing the total loss by the total flow volume for each plot. To compare runoff volumes and P losses from plots on an equal time basis, an interval time of 90 min measured from the onset of rainfall was used as the total runoff event time.
The design of this study was a split-plot, with rate as the main plot treatment in a randomized complete block, and the two-way treatment structure of P source and application method as the sub-plot treatments. Four plots containing manure (M) or fertilizer (F) applied as broadcast (B) or incorporated (I) were randomly distributed within one set that was randomly assigned to either low (L) or high (H) application rate. Each of three blocks representing the three replications were also randomly selected such that the experiment was run one block per day with two rainfall simulations performed in the morning or afternoon for each set. The procedure (proc mixed) in analysis of variance (ANOVA) was used to test whether there were significant differences in TSS and P concentrations and losses in runoff water (SAS Institute, 1991). Main effects included all possible interactions, and random effects were block x rate. Multiple comparisons of TSS and P concentrations and losses in runoff water were performed on the means of the three replicates (blocks) using the Fisher protected least significant difference (LSD) test (Steel and Torrie, 1980).
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Average TSS loss (sediment yield), which is the product of TSS concentration and flow volume, varied from 2.4 to 5.4 Mg ha-1 (Table 2). There was no effect of rate or application method on TSS losses, but manure plots had lower losses than fertilizer plots (P < 0.001). Although TSS concentrations were higher from incorporation plots, the lower flow volumes from these plots resulted in losses that are not significantly different from broadcast plots. Average TSS losses from broadcast manure plots were 45 and 37% lower than broadcast fertilizer plots at high and low rates, respectively; whereas the corresponding losses from incorporation plots were 30 and 12% lower from manure than fertilizer plots. Manure may have acted as the cementing agent for soil aggregates that reduced the impact of rain and lowered the detachment of soil particles during runoff, resulting in lower sediment concentration and yield. Gilley and Eghball (1998) measured TSS of 4.1 Mg ha-1 in a disked fertilizer treatment and 0.9 Mg ha-1 in a disked beef cattle manure treatment. Similarly, Mostaghimi et al. (1992) reported sediment loss in runoff water at 0.48 Mg ha-1 when wastewater sludge was broadcast on a silt loam soil compared with a loss of 5 Mg ha-1 with granular fertilizer. It should be noted that sediment losses observed in our study included also manure or fertilizer particles eroding from the surface; therefore, values of TSS may overestimate true soil loss. Furthermore, these sediment yields depict edge-of-field soil losses, and hence soil moved on a field and not necessarily from a field; the latter is more accurately estimated by a sediment delivery ratio that usually decreases with increasing area (Trimble and Crosson, 2000).
Phosphorus Concentration in Runoff
Flow-weighted concentrations of P forms in runoff water are shown in Table 3. All concentrations far exceeded the eutrophication limit of 0.01 to 0.02 mg L-1 (Vollenweider, 1968) with initial DRP concentrations during the first few minutes of runoff from individual plots reaching as high as 9 mg L-1 (Tabbara, 2000). Mean DRP concentrations in runoff water were lowest from manure incorporated at low rate (0.68 mg L-1) and highest from fertilizer broadcast at high rate (2.58 mg L-1). Dissolved reactive P and TDP concentrations at high application rates followed the order: broadcast fertilizer > broadcast manure > incorporated fertilizer > incorporated manure. Plots in which manure or fertilizer was incorporated had lower DRP and TDP concentrations in runoff water than those from broadcast application (P < 0.001). Incorporating manure reduced average flow-weighted DRP concentration by 58 and 41% and incorporating fertilizer by 41 and 50% at high and low rates, respectively. There was also a difference between manure and fertilizer in DRP, DOP, and TDP concentrations (Table 3). All concentrations were lower from manure than fertilizer when broadcast or incorporated (P < 0.001). This was a result of both a lower rate of P applied and a much lower concentration of soluble P in the applied manure than fertilizer (Table 1).
|
A complicating factor in determining the water quality of surface runoff from agricultural land is the dynamic nature of the system both in the short term during the rainfall-runoff event and in the long term in the times between events. A thin mixing zone of 1 to 5 cm thickness at the soil surface is considered to be the source of DRP (and PP). As DRP is removed from this layer by water moving over and down through the layer, the amount remaining, and therefore the concentration in subsequent rainfall-runoff/leaching waters, decreases. The degree of this decrease is dependent in part on the ability of the soil in this layer to release and replenish the DRP in the water (which in turn depends on kinetic and source-amount factors). This trend with time during a rainfall-runoff event was evident in data from our rainfall simulation study where rainfall intensity was kept constant. Dissolved reactive P concentration in runoff water from fertilized plots displayed a first-order exponential decay during the 90-min rainfall-runoff event with initial concentrations and decay constant both dependent on P source and application method (Fig. 1) .
|
When P amendments are incorporated below the soil surface, P concentration in eroded soil (sediments) is decreased since P is moved out of the mixing zone. This effect seems to outweigh the increase in TSS concentration. It could be concluded that under the conditions of this experiment the source of P in the mixing zone that is available for runoff is more critical in P loss than cultivation practices that would tend to increase soil erosion. This conclusion is augmented by the fact that the storm event was extreme enough that any effect of cultivation on increasing P loss would have shown. The rainfall intensity used was constant and therefore we could not corroborate this conclusion with comparison with a less intense storm. Mueller et al. (1984) compared soil and water losses from a high-intensity storm with a low-intensity storm and found that although significant differences in soil and water losses resulted for both intensities, relative differences among treatments did not vary with rainfall intensity. They suggested that other variables in their experiment, such as residue cover, had greater influences on differences among tillage practices than rainfall application rates. Edwards and Daniel (1993) also studied the effect of rainfall intensity on quality of runoff from liquid swine manure application on grass plots. Their results showed that TP and DRP concentrations in runoff decreased with rainfall intensity (due to dilution), but the higher runoff volumes associated with higher rainfall intensities resulted in comparable mass losses.
Mean BAP concentration, which includes DRP and the portion of PP that might be algal-available, varied in concentration from 1.9 to 13.6 mg L-1 and constituted from 20 to 50% of the TP concentration. Average BAP concentrations were lower from manure than fertilizer for all treatments (Table 3). As for TP and PP, analysis of variance of BAP concentrations in runoff from all plots showed similar and significant effects of rate and application method.
Phosphorus Losses in Runoff
Mean losses varied from 0.4 to 1.7 kg ha-1 as DRP, from 1 to 9.3 kg ha-1 as BAP, and from 5.3 to 24 kg ha-1 as TP, depending on treatment (Table 4). The higher losses are significant from both agronomic and environmental viewpoints. Analysis of variance at the P = 0.05 significance level showed that application rate had a significant effect on TP, PP, and BAP losses but not on soluble P losses. Results also showed that mass losses of all P forms were lower from manure than fertilizer and from incorporated than broadcast plots with significant interaction between method and source of application. Incorporating fertilizer reduced TP loads by 45 to 55%, while incorporating manure reduced TP loads by 8 to 42%, depending on application rate. Bioavailable P mass losses were reduced on incorporation by 62 to 70% for fertilizer and by 42 to 68% for manure. Incorporation reduces surface P concentration and increases the potential for P to bind with soil and thus decreased TP and BAP losses.
|
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
There was no evidence that surface application of manure or fertilizer imparts on the soil surface a protective sealing layer that would reduce losses of P to runoff water. Soluble and sediment-bound P concentrations and losses were much higher with surface application. It appears that manure or fertilizer particles would be saturated when subjected to a heavy rainfall soon after application and thus would not exhibit the short-term "plugging" effect that normally caps the soil surface and prevents nutrient and solid losses (K.A. Smith, personal communication, 2000).
Concentration and losses of all P forms were higher from fertilizer than manure plots. The fraction of applied P lost as TP in runoff water from manure plots was 53% of that from fertilizer plots when broadcast and 89% when incorporated. The higher solubility of fertilizer P than manure P in water leads to higher losses in runoff. Phosphorus fertilizer contained polyphosphate molecules that hydrolyze at a rate dependent on chemical and biological conditions of the soil. These results show that the form of P in manure or fertilizer affected the release and transport of soluble and particulate P in runoff water. It would be interesting to extend future studies to examine the loss of P from this and other "slow release" fertilizers.
|
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:
|
|
 |
|
  B. L. Allen and A. P. Mallarino Effect of Liquid Swine Manure Rate, Incorporation, and Timing of Rainfall on Phosphorus Loss with Surface Runoff J. Environ. Qual., January 4, 2008; 37(1): 125 - 137. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
A. S. Algoazany, P. K. Kalita, G. F. Czapar, and J. K. Mitchell Phosphorus Transport through Subsurface Drainage and Surface Runoff from a Flat Watershed in East Central Illinois, USA J. Environ. Qual., April 5, 2007; 36(3): 681 - 693. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Volf, G. R. Ontkean, D. R. Bennett, D. S. Chanasyk, and J. J. Miller Phosphorus Losses in Simulated Rainfall Runoff from Manured Soils of Alberta J. Environ. Qual., April 5, 2007; 36(3): 730 - 741. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Shigaki, A. Sharpley, and L. I. Prochnow Source-Related Transport of Phosphorus in Surface Runoff J. Environ. Qual., October 27, 2006; 35(6): 2229 - 2235. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. L. Soupir, S. Mostaghimi, and E. R. Yagow Nutrient transport from livestock manure applied to pastureland using phosphorus-based management strategies. J. Environ. Qual., July 1, 2006; 35(4): 1269 - 1278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Ginting and M. Mamo Measuring Runoff-Suspended Solids Using an Improved Turbidometer Method J. Environ. Qual., April 3, 2006; 35(3): 815 - 823. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Little, D. R. Bennett, and J. J. Miller Nutrient and Sediment Losses Under Simulated Rainfall Following Manure Incorporation by Different Methods J. Environ. Qual., September 8, 2005; 34(5): 1883 - 1895. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. C. Daverede, A. N. Kravchenko, R. G. Hoeft, E. D. Nafziger, D. G. Bullock, J. J. Warren, and L. C. Gonzini Phosphorus Runoff from Incorporated and Surface-Applied Liquid Swine Manure and Phosphorus Fertilizer J. Environ. Qual., July 1, 2004; 33(4): 1535 - 1544. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
