|
|
||||||||
Department of Agronomy and Horticulture, 279 Plant Science, University of Nebraska, Lincoln, NE 68583-0915
* Corresponding author (cwortmann2{at}unl.edu)
Received for publication March 7, 2005.
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: BAP(unf), bio-available phosphorus in unfiltered runoff FeO-P, soil P availability using the iron oxide soil P test RP(<0.45), dissolved reactive phosphorus in filtered runoff STP, soil test phosphorus TP, total soil phosphorus TP(unf), total phosphorus in unfiltered runoff WSP, water-soluble phosphorus in soil
| INTRODUCTION |
|---|
|
|
|---|
Phosphorus pollution problems associated with manure are often associated with excessive rates of manure application to cropland. Manure applied to meet crop N need generally results in P application in excess of that removed by the crop. This can result in increased soil P loading and excessive STP levels.
Runoff-P concentration has generally been found to be linearly related to STP, with variation in STP accounting for 60 to 90% of the variation in dissolved P (Sauer et al., 2000; McDowell and Sharpley, 2001; Andraski and Bundy, 2003; Daverede et al., 2003). Median values from these studies for increases in dissolved P in runoff are 3.1 and 2.4 ug L1 for each 1 mg kg1 increase in Mehlich-3 and Bray-1 P, respectively. Klatt et al. (2003) found that P concentration in discharge water from natural runoff events was related to STP, which ranged from 22 to 45 mg kg1 Bray-P1 for the 0- to 15-cm surface soil. They also found that total P in runoff increased by an average of 10.9 and 7.9 µg L1 per milligram-per-kilogram increase in Mehlich-3 and Bray-P1, with R2 values of 0.59 and 0.97, respectively.
Soil test P change points at which the rate of runoff P concentration significantly increases with increasing STP have been reported. Maguire and Sims (2002) found change points at 40 to 50 mg kg1 with Fe oxide soil test P, and at about 200 mg kg1 with Mehlich-3 soil-extractable P. The degree of P saturation as measured by P/(Al + Fe) with Mehlich-3 extraction was more useful than Mehlich-3 alone in identifying a point above which the dissolved reactive P (RP) concentration in runoff greatly increased. This breaking point ratio, 0.2, was greater than the apparent optimum degree of P saturation (0.060.11) for crop production (Sims et al., 2002).
Based on a review of 70 plot-years of data from seven locations under a variety of cropping and tillage conditions, Gilley and Risse (2000) concluded that land application of manure typically decreases water runoff and sediment loss, presumably due to increased water infiltration, and this effect can persist for several years following manure application. While manure application does not always result in reduced runoff and erosion (Gilley and Eghball, 1998), the effect is common enough to be considered as partly offsetting the effect of increased runoff P concentration following manure application.
The increase in STP with manure application is less, per unit of manure applied, if manure P concentration is low. Diet P levels for beef cattle have typically been >0.3% (w/w). It has been shown, however, that beef cattle performance is not diminished by reducing diet P to <0.2% (Erickson et al., 2002) and results in up to 30% less P excretion in manure.
The objectives of this study were to evaluate the effectiveness of soil P tests in predicting P concentration in runoff and to determine the residual effects of low- and high-P composted beef cattle feedlot manure on STP levels, P losses in runoff and sediment, leaching of P, and crop yield.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Low P (0.200.36% P) and high P (0.360.46% P) composted feedlot manure was applied annually to six compost treatments in each replication for 3 yr before the 2001 crop season. The P levels were achieved by feeding beef cattle rations of low (0.240.28%) and high (0.350.45%) P content. Low-P and high-P composts were each applied in three management treatments: winter surface application; spring pre-plant application incorporated by disking; and spring post-plant surface application with no incorporation. A replicated control was included as the seventh treatment, consisting of 200 kg N ha1 applied to corn as NH4NO3 broadcast before incorporation with spring tillage. Tillage for all plots consisted of disk-tillage each year before planting for the years 19982004. Consequently, the spring post-plant treatments remained unincorporated for 1 yr and the winter surface treatments remained unincorporated for
5 mo (January to spring tillage in May). The rate of compost application was calculated to supply 200 kg ha1 plant-available N, assuming that 30% of the organic N would be mineralized during the first cropping season. During the course of 3 yr, a total of 750 and 1150 kg ha1 of P was applied with the low-P and high-P compost, respectively. The last compost applications were in the spring of 2000 except for the winter-applied treatments, where application was made in January of 2001. Plots receiving this last winter application in 2001 were excluded from the analysis of 2001 data, as this study addressed residual effects of manure application.
During the 3-yr period when compost was applied, all plots were planted to corn (Zea mays L.). Beginning in 2001, soybean [Glycine max (L.) Merr.] was planted followed by a cornsoybean rotation sequence. Corn and soybean varieties were glyphosate [N-(phosphonomethyl)glycine] tolerant and weeds were controlled with application of glyphosate as needed. Irrigation water was supplied during the growing season when available water holding capacity reached 50% depletion. All runoff events occurred from March through August and were associated with rainfall or snowmelt events, except for one event in July 2002, which was induced with excessive irrigation.
The area harvested for grain yield determination was two rows of 9.0-m length. Corn was hand harvested and shelled after drying. Soybean was machine harvested. Grain P concentration was determined by x-ray fluorescence spectroscopy of solid plant material (Knudsen et al., 1981). Grain N concentration was determined using a modification of the Dumas method (Bremner, 1996).
Soil and Water Sample Collection and Analysis
The surface soil was sampled in the spring of 2004 for the 0- to 5-, 5- to 10-, 10- to 15-, and 15- to 30-cm depths. Composite samples were formed for each plot of 10 cores of 19.1-mm diameter for the 0- to 15-cm depths and 17.5-mm diameter for the 15- to 30-cm depth.
Surface soil samples were analyzed for total P (TP) by perchloric digestion, extractable P by the Bray-P1, Mehlich-3, and Olsen tests (Kuo, 1996), and FeO-P according to Chardon (2000). Water-soluble soil P (WSP) was determined by the method described by Self-Davis et al. (2000). Phosphorus concentration in soil extracts was colorimetrically determined by the molybdate blue method (Murphy and Riley, 1962). Soil particle size distribution was determined using a sieving and sedimentation procedure (Kettler et al., 2001). Bulk density was derived for the 0- to 5- and 5- to 10-cm depths from five cores per plot collected with a 19.1-mm-diameter soil probe. Soil organic matter was determined by loss on ignition (Nelson and Sommers, 1996).
The volume of runoff was determined after each runoff event by measuring the depth of collected runoff in the tank and calculating the volume using equations developed for this determination. A 1-L subsample was taken and stored at 4°C until analysis was performed during the winter. Sediment concentration was gravimetrically determined by drying 10 mL of unfiltered runoff sample at 104°C. Algae-available or bio-available P for unfiltered samples [BAP(unf)] was determined by extraction with FeO-impregnated filter paper (Myers and Pierzynski, 2000). Total P in unfiltered runoff samples [TP(unf)] was determined using HClO4HNO3. Dissolved reactive P in runoff was determined after <0.45-µm filtration of a 100-mL sample [RP(<0.45)] (Pote and Daniel, 2000). In all cases, runoff P concentrations were measured colorimetrically according to Murphy and Riley (1962).
Statistical Analysis
Results were analyzed using Statistix 8 (Analytical Software, 2003) and analysis of variance was conducted by year for all measured data. The residual effects of previously applied compost did not differ due to time and method of application, so data were pooled across these treatments (the winter application treatments were excluded in 2001) and statistically analyzed for the main effects of compost (n = 18) vs. no compost (n = 3) and of high-P (n = 9) vs. low-P (n = 9) compost treatments. Runoff volumes were highly variable within individual runoff events and the analysis of variance was run for the totals of runoff volume, sediment, and nutrient loss within each year and mean annual runoff concentrations, [P]
, which were calculated from these totals for each plot as follows:
![]() | [1] |
![]() | [2] |
Soil test procedures were compared using Pearson correlation and linear regression analysis, and PROC NLIN (SAS Institute, 2000) for split-slope analysis, using the individual observations for the soil samples collected in 2004. The relationship of STP in 2004 for the 0- to 5-cm depth was related to mean runoff P concentration using regression analysis. Mean runoff P concentration for these regressions was determined by dividing total mass of runoff P lost by the total volume of runoff for the 20022004.
| RESULTS |
|---|
|
|
|---|
|
|
Correlation among Soil Phosphorus Tests
The results of the various soil P analyses for samples collected from the different treatments and depths were highly correlated (r > 0.97) with the exception of WSP, which was weakly correlated (r = 0.450.53) with all other tests. The correlation of Bray-P1 with Mehlich-3 and Olsen P was very strong, but weakened after Bray-P1 exceeded 400 mg kg1. When Bray-P1 was <400 mg kg1, Bray-P1 = 17.05 + 0.882 Mehlich-3 (R2 = 0.96, P < 0.001) and Bray-P1 = 6.46 + 0.450 Olsen P (R2 = 0.95, P < 0.001). Bray-P1, Olsen P, and Mehlich-3 P increased as TP increased. Bray-P1 = 164.14 + 0.420 total P (R2 = 0.96, P < 0.001), Mehlich-3 P = 164.04 + 0.372 total P (R2 = 0.93, P < 0.001), and Olsen P = 69.95 + 0.191 total P (R2 = 0.94, P < 0.001).
The percentage of TP that was extractable by agronomic soil P tests increased as TP increased to approximately 900 mg P kg1 (Table 3).
|
Runoff and Phosphorus Losses
Total runoff volume during the 3-yr period following compost application was 41% less than for plots that did not receive compost application (Table 4). With the greater runoff volume, sediment loss was 120% greater (by weight) where no compost had been applied than the mean loss for the compost treatments. Runoff volume and sediment loss were not affected by compost type. Particle size distribution (723% sand) and slope (4.07.5%) of the individual plots were not related to total runoff volume and sediment loss. Sediment concentration was generally low, as might be expected given that the effective slope length of 11 m was too short for the runoff to have much erosive power in most runoff events. The decreased runoff and sediment loss with compost application presumably was due to increased water holding capacity (higher soil organic matter), soil aggregation, and rate of water infiltration associated with compost-amended soil. These results are in agreement with the findings of Gilley and Risse (2000). McDowell and Sharpley (2003) observed increased soil aggregation, decreased slaking, and reduced sediment following manure application, which they attributed to added labile soil C.
|
|
Soil Tests and Runoff Phosphorus Concentration
Soil test P was linearly or quadratically related to the three species of runoff P for all soil tests (Table 6). The amount of variation in runoff P concentration explained by the soil tests was greatest for Bray-P1 and Mehlich-3 and least for FeO and WSP. The BAP(unf) concentration was generally best accounted for by the soil tests. Klatt et al. (2003) found, however, that FeO-P, Mehlich-3 P, and Olsen P predicted runoff P concentrations with similar accuracy but better than Bray-P1.
|
Runoff P concentrations were already substantial at STP levels considered moderate for optimal crop growth (Table 5) as indicated by the positive y intercept values (Table 6). Total P(unf) loss with the no-compost treatment was 56% of that with the compost treatments, despite large differences in STP (Table 2). The y intercepts are always positive, except for one quadratic equation, and high relative to those reported by others (McDowell and Sharpley, 2001; Andraski and Bundy, 2003; Daverede et al., 2003; Klatt et al., 2003). The high y intercepts are supported by other reports of substantial runoff P concentrations at agronomically moderate STP levels (Eghball et al., 2000; Eghball and Gilley 1999).
Agronomic soil test P needs to be increased by more than 140, 500, and 500% to double RP(< 0.45), BAP(unf), and TP(unf) concentrations, respectively (Table 6). Our results indicate that increases in STP due to manure application are of much less importance than P transport factors, including runoff and erosion, to P loss in runoff. In comparing the no-compost treatment to the mean of the high-P compost treatment, the respective changes in Bray-P1, runoff volume, sediment loss, and total P loss in runoff were 4900, 49, 50, and 130%. The very large difference in STP resulted in increased P loss, which was partly compensated by reduced runoff and erosion to the extent that there were no treatment effects on total P loss in 2 of 4 yr. When given a choice of building soil P to excessive levels on a field with little runoff and erosion potential vs. spreading manure P more widely on land with moderate to high runoff and erosion potential, the first option would be the most environmentally sound.
These results are relevant to improvement of P-loss risk assessment tools. Many P indices are very sensitive to changes in STP and give very low scores when available soil P is of very low to medium availability (Benning and Wortmann, 2005). Our results show that P risk sensitivity to transport factors is more important than STP. In some P indices, the score for the STP factor, but not the P index score, which can continue to increase due to other factors, is maximized at a certain STP level, such as 120 or 150 mg kg1. For such P indices, the STP factor score should be allowed to increase indefinitely as STP increases.
Leaching of Phosphorus
Bray-P1, Mehlich-3 P, Olsen P, and FeO-P were greater below the tillage depth (>15 cm) with compost than no compost, suggesting downward movement of P (Table 2). Differences in WSP and TP at these depths, however, were not significant. There is no evidence of significant P leaching below the 30-cm depth, as Bray-P1 for the 30- to 60-cm depth was not significantly different between compost and no-compost treatments (data not shown). In a simulation study, Sharpley and Moyer (2000) found significant leaching of P related to the amount of manure P applied and to WSP in the soil following manure application. They found that organic P was more readily leached than inorganic P. These results differ from our findings, where applied P had more time to react with soilprobably with much of it embedded in soil aggregatesto the extent that WSP was similar for the no-compost and compost treatments.
Crop Yields
Corn yield was not affected by compost treatment but soybean yield was increased where compost had been applied (Table 7). Grain N concentration was not affected by treatments and only soybean grain N content in 2003 was higher with low-P compost applied than with the other treatments. Soybean grain P concentration and content were similar for low- and high-P compost, but corn grain P concentration and content were less in the low-P compost than the high-P compost treatments. Grain P concentration and content were always less for the no-compost treatment than for the compost treatments. This may partly account for the lower yield with the no-compost treatment, as its STP was less than considered optimum for corn and soybean performance. Total P removal by crops during this 3-yr period amounted to <9% of total P applied in compost.
|
| CONCLUSIONS |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
| NOTES |
|---|
|
|
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. S. Wortmann and D. T. Walters Residual Effects of Compost and Plowing on Phosphorus and Sediment in Runoff J. Environ. Qual., August 31, 2007; 36(5): 1521 - 1527. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Garcia, C. S. Wortmann, M. Mamo, R. Drijber, and D. Tarkalson One-Time Tillage of No-Till: Effects on Nutrients, Mycorrhizae, and Phosphorus Uptake Agron. J., June 26, 2007; 99(4): 1093 - 1103. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. A. Quincke, C. S. Wortmann, M. Mamo, T. Franti, R. A. Drijber, and J. P. Garcia One-Time Tillage of No-Till Systems: Soil Physical Properties, Phosphorus Runoff, and Crop Yield Agron. J., June 26, 2007; 99(4): 1104 - 1110. [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 |