| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a AgResearch Ltd., Invermay Agricultural Centre, Private Bag 50034, Mosgiel, New Zealand
b Pasture Systems and Watershed Management Research Unit, Curtin Road, University Park, PA 16802-3702
* Corresponding author (richard.mcdowell{at}agresearch.co.nz)
Received for publication February 19, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: CEC, cation exchange capacity • CHO, carbohydrate • DRP, dissolved reactive phosphorus • DURP, dissolved unreactive phosphorus • PP, particulate phosphorus • TDP, total dissolved phosphorus • TP, total phosphorus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Land application of manure to increase fertility also affects soil structure and erosion potential through the addition of organic matter and cations such as sodium and calcium (Mamedov and Levy, 2001). For example, when a soil previously in grassland is cultivated for arable cropping, soil organic matter and total carbon (C) decrease along with aggregate stability, which increases the potential for erosion (Sharpley and Smith, 1983). However, changes in soil aggregate stability can occur without much change in total C. For example, Haynes and Swift (1990) showed that a marked difference in aggregate stability between a regrassed soil (13-yr arable and 2-yr grassland) and a grassland soil (15-yr grassland) was closely correlated with hot water–extractable carbohydrate and not total C.
When a soil is plowed and exposed to the air, changes in organic matter quality are also accompanied with a decrease in soil moisture and an increase in the soil's potential for erosion processes such as slaking and dispersion. Recent work has shown that during overland flow, an air-dry soil (2% moisture) that is suddenly brought into contact with high-intensity rainfall (65 mm h-1) loses much more finer-sized particles and P (via selective erosion) at the beginning than at the end of an event (McDowell and Sharpley, 2002).
While the application of some manure can have beneficial effects for soil structure and in turn decrease the potential for erosion, the overapplication of manure can also increase the potential for P loss in overland flow (McDowell et al., 2001c; Sharpley et al., 1998). The addition of inorganic P fertilizers to a soil may lack some of the structural benefits imparted by manure, while still posing a threat to water quality (Sharpley and Menzel, 1987).
In continental and/or humid climates, the loss of topsoil and associated P is one of the major factors associated with decreasing crop yields and degrading water quality largely via short high-intensity storms (Pionke et al., 2000). The loss of P in watersheds with a large amount of intensive agriculture and/or manure application to land is often associated with the loss of a considerable amounts of C in dissolved and particulate forms (Fleming and Cox, 2001). Clearly, the response of the soil (manured or not) to intense storms that erode topsoil is vital for the prediction of potential P transfer. Thus, we hypothesize that by considering manure management, and soil chemistry such as exchange cations and C quality and quantity (and, thereby, resistance to slaking and dispersion processes), we may better account for potential P loss during overland flow. To test this hypothesis, we analyzed the loss of P in different fractions (dissolved and particulate), sediment, and C in overland flow with time from soils that received P as manure or fertilizer.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soil Analyses
Soils from plots of each treatment were analyzed in triplicate. This included Mehlich 3–extractable P (Mehlich, 1984), water-extractable P (1 g soil 100 mL water), total P (Taylor, 2000), and total C (soils ground to <250 µm; determined with a LECO [St. Joseph, MI] C/N analyzer). Labile carbohydrate was determined by autoclaving 2 g of soil with 20 mL of 0.5 M H2SO4 at 103 kPa for 1 h at a temperature of approximately 121°C (Lowe, 1993). The suspension was filtered and labile carbohydrate analyzed colorimetrically with anthrone method with glucose standards (Cheshire, 1979). A sequential extraction procedure was used to extract carbohydrates from soils. Soils (2 g each) were sequentially extracted with 20 mL of cold water, hot (353 K) water, 1 M HCl, and 0.5 M NaOH. Samples were centrifuged (2600 x g) for 10 min to prevent soil loss before adding the next extractant. Each supernatant was filtered to <0.45 µm before colorimetric analysis with anthrone. Exchangeable Al, Ca, K, Mg, and Na and cation exchange capacity were determined with the ammonium acetate procedure outlined by Hendershot et al. (1993). The exchangeable sodium percentage (ESP) was calculated as the quotient of exchangeable Na and cation exchange capacity (CEC) x 100%.
Particle size distribution was determined by hydrometer, following dispersion of the samples with sodium hexametaphosphate by shaking for 24 h. Undispersed samples were analyzed after shaking with water for 10 min. The degree of soil aggregation is represented as the ratio of the proportions of clay-sized materials (<2 µm) in dispersed and undispersed soil.
Overland Flow Study
Duplicates of air-dry soil from each plot were packed into boxes (15 cm wide, 10 cm deep, and 100 cm long with six, 3-mm-diameter holes drilled in the base to allow some drainage) to a bulk density of 1.2 g cm-3 (equivalent of field conditions). Saturation-excess overland flow was generated by applying rainfall (tap water, P less than detection limit of 0.005 mg P L-1) at 65 mm h-1 to each boxed soil, set at a five percent slope. This rainfall intensity and duration has an approximate 5-yr return frequency in central Pennsylvania. All rainfall was produced with precalibrated nozzles 2.7 m above the soil surface generating raindrops with size, velocity, and impact angles approximating natural rainfall (Shelton et al., 1985). Samples of flow were taken every 5 min after the onset of flow for the first 30 min and every 10 min thereafter for an additional 30 min.
Water Analyses
Samples collected at each time (5, 10, 15, 20, 25, 30, 40, 50, and 60 min after the onset of flow) were filtered (<0.45 µm) and stored at 4°C in the dark until analysis. An unfiltered sample was also kept. All P measurements were made colorimetrically and in duplicate (Murphy and Riley, 1962). Within 24 h, each sample was analyzed for dissolved (<0.45 µm) reactive phosphorus (DRP), and within 48 h for total dissolved reactive phosphorus (TDP) and total phosphorus (TP) following a Kjeldahl digestion (Taylor, 2000). Fractions defined as dissolved unreactive phosphorus (DURP) were determined as the difference between TDP and DRP and particulate phosphorus (PP) as TP less TDP. Suspended sediment was collected and determined after 7 d, by centrifugation (2600 x g), removal of the supernatant, and air-drying the residue.
All statistical analyses (mean and correlation coefficients) were performed with SPSS Version 10.0 (SPSS, 1999). For the comparison of means a one-way analysis of variance was conducted in a randomized design. However, due to the use of pseudo-replication (duplicates were used of each plot, themselves only duplicated), no additional analyses were performed. As a result, caution should be employed when interpreting data from individual plots. All correlation coefficients were generated with linear functions except when power functions exhibited a better fit to the data.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
When rainfall (65 mm h-1) was applied to the untreated and dairy-manured Hagerstown soils, dairy manure–treated soils (200 kg P ha-1) with 32.8 g C kg-1 took twice as long (21 min) to start flowing as untreated soil (10 min) with only 16.3 g C kg-1 (Table 2). No consistent trends in time to flow initiation were seen for soils treated with poultry manure or superphosphate, reflecting little change in total soil C for these soils (Table 1).
|
Soils that have a greater organic matter concentration are also more resistant to aggregate breakdown, slaking, and dispersion than soils poor in organic matter. Similarly, increasing exchangeable Ca (and to some extent K; Chen et al., 1983) decreases dispersion, while an increase in Na will increase dispersion (Kay and Angers, 2000). Our soils contained very little exchangeable Na (all exchangeable sodium percentage values < 1%), but varied quite considerably in their Ca concentration (4.2 to 11.5 cmol kg-1; Table 1). Unsurprisingly, a clear negative correlation was found between CEC and the load of sediment lost from each soil (sediment lost = 76 x CEC-0.77, r = -0.56, P < 0.05) and similarly between exchangeable Ca and sediment lost (sediment lost = 79 x exchangeable Ca-0.91, r = -0.59, P < 0.05).
Previous work has also shown that the effect of organic matter depends not only upon quantity, but also upon C quality, specifically soil carbohydrate concentration (e.g., Cheshire, 1979; Haynes, 1993; Tisdall and Oades, 1982). Therefore we sequentially extracted different pools of soil carbohydrate (CHO, expressed relative to glucose) (Table 3). A stronger negative relationship was obtained for hot water (HWCHO)– and NaOH-extractable carbohydrate (NaOHCHO) to sediment lost than with total C concentration (sediment lost = 8.7 x HWCHO-0.42, r = -0.60, P < 0.05; sediment lost = 64 x NaOHCHO-1.48, r = -0.55, P < 0.05; sediment lost = 412 x total C-1.15, r = -0.54, P < 0.05; Tables 3 and 4). No significant relationships were obtained between cold water–extractable or acid-extractable carbohydrate and total sediment loss.
|
|
Our data also showed that the degree of soil aggregation within the soil was negatively correlated to sediment lost in overland flow (sediment lost = 30 x degree of soil aggregation-0.62, r = -0.51, P < 0.05; Tables 1 and 4). However, the strongest negative correlation to total sediment transported was found with acid-hydrolyzable carbohydrate (termed here labile carbohydrate; r = -0.62; Fig. 1 and Tables 3 and 4). In addition, this fraction (not part of the sequential extraction regime) was more strongly positively correlated to the degree of soil aggregation than sequentially extracted fractions (degree of soil aggregation = 4.34 x labile carbohydrate-2.91, r = 0.81, P < 0.01; other sequentially extracted fractions, r = 0.27, 0.44, 0.13, and 0.37 for cold waterCHO, HWCHO, HClCHO, and NaOHCHO, respectively; Tables 3 and 4). Thus, the labile carbohydrate pool appears to best approximate carbohydrates active in determining soil binding (degree of aggregation) and to some extent sediment transport. However, a better positive correlation was obtained with CEC (r = 0.70, P < 0.05), and more so by considering exchangeable Ca, alone (r = 0.72, P < 0.05); indicating a relatively strong role of Ca in sediment transport.
|
Variation of Phosphorus Transport during Overland Flow
The volume overland flow was similar (P < 0.05) for all treatments (data not presented). Therefore, P transport in overland flow is a function of soil P fractions, which in turn reflect the rates and form of P applied. To describe the pattern of P concentrations for each P fraction with time after the onset of saturation-excess flow, we fitted a power function to the data:
 | [1] |
and ß were constants related to the initial concentration of, and decrease with flow time of P fractions, respectively. Looking at parameter , values for the dissolved fractions (DRP and DURP) increased with the application of each amendment (Table 5). However, the same was not true of TP, largely due to the contribution of PP and greater influence of physical processes on P transport (e.g., detachment of sediments or flocs, slaking, and dispersion).
|
In general, concentrations of each P fraction decreased with time (ß parameter negative, Eq. [1]; Fig. 2 and Table 5). Sharpley (1980) attributed a 50% decrease in DRP concentrations with time to the dilution of soil solution. While this may occur in systems where little sediment is lost, it is an oversimplification where erosion processes are also involved. Stoltenberg and White (1953) found that fine material is selectively removed from the soil surface, which latter develops a crust or "pavement" of coarser material at the surface. In general, fine particles contain much more sorptive surfaces and P than coarser particles (Sharpley, 1985). Data in Table 6 indicates that in general more P was lost in sediment from the first 30 min of flow than during the second 30 min, presumably due to selective erosion of fine particles rich with P. The erosion of P-rich particles influences P fractions via a pseudo-equilibrium, whereby P fractions reflect the concentration and availability of P-rich particles in overland flow (Fig. 2).
|
|
Figure 3 shows that the parameter ß for DRP and TP, which describes the pattern of P concentration with time, is significantly correlated with the concentration of labile carbohydrate, such that a high concentration of carbohydrate yields a small ß value. However, no such correlation exists between the parameter ß for DRP and TP and exchangeable Ca. This infers that during the experiment, the pattern (but not necessarily the load) of P loss was a function of soil particle slaking and that clay dispersion was less involved. However, we do not wish to imply that dispersion is not occurring. Moreover, the pattern of P loss is defined by those particles that release and contain the most P and relates to their physical binding by carbohydrates and availability in flow, which conforms to the concept of P movement in carbohydrate-rich flocs.
|
|
Soil Carbon and Phosphorus Interactions and Perspective for Management
In manured soils it is nearly impossible to separate the effects of C and P; this is also difficult in soils that have received inorganic fertilizers due to the concomitant increase in plant production and, in turn, soil C reserves. The data presented here show that soil carbohydrates play a role in determining how well P is exposed to overland flow and, thus, the potential for loss. However, we do not imply that manure application at any rate will have this effect. In contrast, the effect upon soil physical characteristics is clearly limited beyond a certain "critical" application rate. Manure application in excess of this rate will result in excessive P loss in overland flow in relation to unmanured soil or soil manured at a rate less than the critical rate.
Previous work has shown some evidence for the formation of low-density flocs, whereby manure and soil form a low-density particle containing much P that is preferentially lost over more dense soil-only particles (Droppo et al., 1998; McDowell et al., 2001a). Thus, the "critical" rate of manure application will be influenced by the concentration of labile carbohydrates and availability (fractionation) of P in manure. For example, more DRP was transported from soil receiving 200 kg P ha-1 as dairy manure (3.33 mg) than poultry manure (1.39 mg) (Table 4). In contrast, more TP was transported from the poultry-manured soil (32.1 mg) than dairy-manured soil (7 mg), largely due to the greater contribution of PP (Table 4).
Sharpley and Moyer (2000) showed that on a manure-only basis, poultry manure should leach more P (as inorganic P) than dairy manure. However, this study and recent evidence presented by McDowell and Sharpley (2002) suggest the opposite when manure is mixed with soil. This is due to the differences imparted by the component of the manure that is not P (e.g., organic matter and carbohydrates). Table 3 shows that dairy-manured soils contain much more labile carbohydrates than poultry-manured soils receiving the same P application rate. Thus, more protection is gained against slaking and dispersion, which in turn facilitates the transport of P as PP.
We remind the reader that this study is conducted for only one soil type and as such caution should be employed when relating it to others. However, the data does demonstrate some of the processes controlling P loss and their interaction with C quality. Although insufficient replicates were available to make statistical comparisons among manure rates (only manure types), we can infer from the data that the rate of manure application affects P fractionation and load. The benefit of manure application is evident in the relationship between labile carbohydrate or exchangeable Ca and sediment loss (Fig. 1). Labile carbohydrate had a significant effect upon the pattern and load of TP lost, and along with exchangeable Ca also affected the load of TP lost. For example, untreated soil lost more P as TP than soils that had received dairy manure at a rate of 25 or 50 kg P ha-1. Beyond this rate, the physical soil benefits are negated by increased P loss (Table 4). These rates are relevant to both agronomic and environmental perspectives.
From a crop production standpoint, supplemental P application would be recommended to the untreated soil (Mehlich 3–extractable P of 19 mg kg-1). Soils receiving 25 and 50 kg P ha-1 as dairy manure are at an optimal plant-available P concentration (Mehlich 3–extractable P between 30 and 50 mg kg-1; Beegle, 1999) and have a lower potential to lose TP than the untreated soil, due to less erosion afforded by increased soil aggregation from added manure. For poultry manure the physical benefit is not apparent (sediment and TP lost > untreated soil) except in the 100 kg P ha-1 manured soil where sediment and TP loss is less than the untreated soil, perhaps due to a greater labile carbohydrate concentration (labileCHO = 1.61 g kg-1 compared with 1.11 g kg-1 in untreated soil; Table 3). However, DRP load from this soil is much greater than the untreated soil.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The effect of manure (especially dairy manure) upon soil P release was more pronounced in soils that had received P as superphosphate. The physical benefits imparted by manure application at a low application rate (
50 kg P ha-1) meant that for this soil type and manure, less TP was lost than an unmanured control soil. However, this benefit is overcome at greater manure application rates.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. N. Sharpley, P. J. A. Kleinman, A. L. Heathwaite, W. J. Gburek, G. J. Folmar, and J. P. Schmidt Phosphorus Loss from an Agricultural Watershed as a Function of Storm Size J. Environ. Qual., March 1, 2008; 37(2): 362 - 368. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R.F. Alleoni, S. R. Brinton, and G. A. O'Connor Runoff and Leachate Losses of Phosphorus in a Sandy Spodosol Amended with Biosolids J. Environ. Qual., January 4, 2008; 37(1): 259 - 265. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Spargo, G. K. Evanylo, and M. M. Alley Repeated Compost Application Effects on Phosphorus Runoff in the Virginia Piedmont J. Environ. Qual., October 27, 2006; 35(6): 2342 - 2351. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. S. Wortmann and D. T. Walters Phosphorus Runoff during Four Years following Composted Manure Application. J. Environ. Qual., March 1, 2006; 35(2): 651 - 657. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. A. Elliott, R. C. Brandt, and G. A. O'Connor Runoff Phosphorus Losses from Surface-Applied Biosolids J. Environ. Qual., August 9, 2005; 34(5): 1632 - 1639. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Udawatta, P. P. Motavalli, and H. E. Garrett Phosphorus Loss and Runoff Characteristics in Three Adjacent Agricultural Watersheds with Claypan Soils J. Environ. Qual., September 1, 2004; 33(5): 1709 - 1719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Shober and J. T. Sims Phosphorus Restrictions for Land Application of Biosolids: Current Status and Future Trends J. Environ. Qual., November 1, 2003; 32(6): 1955 - 1964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. A. Kleinman and A. N. Sharpley Effect of Broadcast Manure on Runoff Phosphorus Concentrations over Successive Rainfall Events J. Environ. Qual., May 1, 2003; 32(3): 1072 - 1081. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Soil Science Society of America Journal | Journal of Plant Registrations | The Plant Genome | |||
