| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
a Dep. of Horticulture and Landscape Architecture, Colorado State Univ., Fort Collins, CO 80523-1173
b Dep. of Soil & Crop Sciences, Colorado State Univ., Fort Collins, CO 80523-1170
* Corresponding author (Jessica.Davis{at}Colostate.edu)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AB-DTPA, ammonium bicarbonate-diethylenetriaminepentaacetic acid • AWC, available water content • CEC, cation exchange capacity • Db, bulk density • EC, electrical conductivity • Kfs, field-saturated hydraulic conductivity • Ks, saturated hydraulic conductivity • OM, organic matter • OP, ortho-phosphate • TDP, total dissolved P • TIN, total inorganic nitrogen • TN, total nitrogen • TP, total phosphorus • 
, volumetric water content
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Soil physical properties are considered a critical component of soil quality (Kutilek, 2004), and research has shown that addition of organic waste materials can greatly improve a soil's physical properties and fertility levels (Epstein, 1975). Some of the beneficial effects organic materials have on soil physical properties include increased water holding capacity, soil aggregation, soil aeration and permeability, and decreased soil crusting and bulk density (USDA, 1957, 1978). Regular additions of organic materials such as animal manures and crop residues maintain the tilth, fertility, and productivity of agricultural soils (Hornick and Parr, 1987). Another important soil parameter is saturated hydraulic conductivity, which measures the ability of a soil to transmit water (Wu et al., 1999). Compost and manure have been shown to increase saturated hydraulic conductivity when incorporated into the soil (Celik et al., 2004).
In addition to impacting soil physical properties, organic materials also affect soil chemical properties. Animal manure has long been used as a source of plant nutrients and organic matter to improve fertility conditions of agricultural lands (Dao and Cavigelli, 2003). Schlegel (1992) showed that soil P, K and organic matter (OM) increased linearly with increased rates of composted manure, and concluded that amending soil with composted manure is effective for maintaining or increasing soil nutrient levels, especially P, without excessive accumulation of NO3–N. Likewise, Cuevas et al. (2000) explained that available P and K, concentration of NO3–N, and EC increased significantly after a composted municipal solid waste application. Aggelides and Londra (2000) found that amending compost into the soil directly affected soil properties, increasing both OM and cation exchange capacity (CEC). Another study conducted by Egashira et al. (2003) showed that OM and N contents of soils treated with compost and cow dung were significantly higher than the untreated control, and compost increased the CEC of the soil as well.
In Colorado, nutrient runoff and leaching has become a major problem along the South Platte River Basin. About 300000 Mg of N and 40000 Mg of P enter the South Platte River Basin annually from wastewater treatment plants, precipitation, fertilizer, and manure (U.S. Geological Survey [USGS], 1995). Of these sources, fertilizer and manure are predominately applied to agricultural land and turfgrass areas as a means to enhance plant growth. In addition to runoff, NO3–N leaching can also be a problem due to its high mobility, especially in sandy soils. The USDA-ARS has reported that nitrate is leaching into the South Platte's shallow alluvial aquifer from fertilizer and manure applications (Schuff, 1992).
Many studies have shown that turfgrass is an excellent system for minimizing nutrient runoff and leaching. The dense growth habit of Kentucky bluegrass and the thick thatch layer reduce the impact energy of raindrops, runoff velocity, soil detachment and sediment loss, while promoting infiltration in a manner similar to a grass buffer strip (Linde et al., 1995, 1998; Easton and Petrovic, 2004). In Maryland, it was found that runoff losses of sediment and nutrients from turf were extremely low when compared with agronomic row crops (Gross et al. 1990, 1991). Linde and Watschke (1997) reported that N and P runoff concentrations from turf areas, under normal soil moisture conditions, were below the EPA drinking water standard. Morton et al. (1988) investigated N runoff from Kentucky bluegrass turf in Rhode Island and reported that concentrations of inorganic N for all the treatments ranged from 1.1 to 4.2 mg L–1, well below the EPA standard. Starr and DeRoo (1981) observed low NO3–N leaching (0.3–10 mg L–1) beneath a Kentucky bluegrass/red fescue lawn over a 3-yr period, and noted that water samples from wells upstream, 25 and 50 m away, averaged 0.9 and 2.7 mg L–1 NO3–N, respectively. Mancino and Troll (1990) investigated NO3 leaching from creeping bentgrass under conditions favoring heavy leaching losses and found that NO3 leaching averaged <0.5% of the applied N at a rate of 9.8 kg N ha–1 wk–1. Miltner et al. (1996) studied the fate of urea applied to Kentucky bluegrass using 15N labeled (NH4) 2SO4 and found that nitrate concentrations in leachate were generally below 1 mg NO3–N L–1, collected in the drainage water of lysimeters. Morton et al. (1988) and Starr and DeRoo (1981) concluded that under management practices common to home lawns, the risk of ground water contamination from fertilizer N is extremely low.
Leaching losses are influenced by soil texture, N management, and irrigation practices (Petrovic, 1990). For example, Morton et al. (1988) showed that overwatering turf on a sandy loam resulted in >90% of N applied being lost to leaching; however, soil water concentrations were well below the U.S. drinking water standard. In addition, Kopp and Guillard (2005) demonstrated that either excessive N fertilization or overwatering bentgrass on a fine sandy loam increased leachate NO3–N concentrations. On the other hand, Geron et al. (1993) evaluated the impact of N source, N rate, and fertilizer scheduling programs on a silt loam and found no differences in NO3–N leaching losses. Again, concentrations in leachate were well below the drinking water standard. Generally, Petrovic (1990) reported that N leaching losses were <10% of N applied.
The turfgrass canopy reduces erosion by dissipating sediment detachment (Krenitsky et al., 1998), and reducing subsequent transport of sorbed ions such as P (Easton and Petrovic, 2004). Gaudreau et al. (2002) examined P concentrations in runoff from newly established bermudagrass broadcasted with composted dairy manure (50 and 100 kg P ha–1) and inorganic fertilizer (25 and 50 kg P ha–1) on an 8.5% slope. They showed that runoff losses of dissolved P from eight rain events were 44% less for composted manure than fertilizer treatment at equal P rates. The authors concluded that P in compost is less soluble and transportable than fertilizer P; however, dissolved P concentrations of runoff from compost treatments were well above 2 mg L–1, which raises some environmental concerns. Nitrogen runoff from turfgrass seldom occurs, based on a literature review by Petrovic (1990).
Little information is available in the literature about effects that topdressing compost onto established turfgrass has on soil properties or water quality. It was hypothesized that topdressing compost on turf decreases soil bulk density and improves hydraulic properties and increases nutrient and OM levels in the soil, without increasing N and P in runoff or below the rootzone in clay loam soil.
The objectives of this study were to evaluate the effects that topdressing composted manure onto established Kentucky bluegrass has on: (i) soil physical properties, (ii) soil chemical properties, (iii) soil NO3–N and P concentrations below the rootzone, (iv) total runoff and sediment losses, and (v) N and P concentrations in runoff.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The plots were core aerated the day before each compost application using a Toro Greens Aerator (Model 09120, Toro Co., Bloomington, MN) with 5.7 cm long tines spaced 6.4 cm apart. Following core aeration, four compost treatments (0, 33, 66, and 99 m3 ha–1) were randomly assigned. When topdressing compost onto the turfgrass, a known volume of compost was spread onto the turf surface as evenly as possible, then swept into the grass and aeration holes with a broom. The compost was produced from organic dairy cattle manure and had a density of 958.7 kg m–3, pH of 9.2, EC of 32.8 dS m–1 (saturated paste), 0.65% N with 272 mg NH4–N kg–1 and 12 mg NO3–N kg–1, 0.27% P, and 1.44% K. Treatments were applied in May 2003, September 2003, and May 2004.
From the beginning of May through the end of September 2003 and 2004, irrigation was applied twice weekly, delivering 4.1 cm of water per week. Electrical conductivity and sodium adsorption ratio (SAR) of the irrigation water were 2.8 dS m–1 and 1.8, respectively. Plots were mowed once a week during the growing season at 6.4 cm height. Clippings were collected four times per growing season to determine clipping yield; however, the rest of the time, the clippings were recycled back into the turf.
To balance treatments in total inorganic nitrogen (TIN) applied, urea (46–0-0) was applied at rates of 62, 39, and 19 kg N ha–1 yr–1 to treatments 0, 33, and 66 m3 ha–1, respectively. Table 1 represents all sources of N applied during the experiment and shows TIN and P applied. The only P or K source applied was the compost itself; no attempt was made to balance application rates of any nutrients except for N.
|
3 cm in height) were collected directly beneath the thatch layer in each plot, while avoiding the aeration core holes, and cores were carefully transferred into PVC rings. Cores were collected with a manually operated drop hammer sampler. Pressure plate apparatuses were used to determine the volumetric water content () of the soil samples at pressures of 10, 30, 200, 500, and 1500 KPa (Soil Moisture Equipment Corp., Santa Barbara, CA). Saturated water content, available water content (AWC), and bulk density (Db) were determined using the same soil cores.
A single ring infiltrometer was used to measure field-saturated hydraulic conductivity (Kfs) based on the methods of Reynolds et al. (2002) on each plot. In summary, 4 to 6 h before measurement initiation, plots were irrigated with 2 cm of water, to speed up the saturation process. An 8.5-cm diam. metal ring was inserted 6.5 cm into the soil in two locations per plot. Two ponding depths of 5 and 10 cm were used to measure the "quasi-steady flow rate," which is the rate at which discharge becomes effectively constant. After steady state was reached, flow rate was measured every minute for the next 10 min.
The formula used to calculate field-saturated hydraulic conductivity (Reynolds et al., 2002) was
 | [1] |
Q represents the change in quasi-steady state flow rate in cm3 s–1; and H is the difference of two pond depths in centimeters. To examine the effects that topdressing compost has on soil chemical properties, soil samples were collected in September 2004 to evaluate treatment effects. Four soil cores (0- to 50-cm depth) were collected from each plot using a Giddings Hydraulic Soil Probe (#15-SCS Model GSRPS, Giddings Machine Company Inc., Windsor, CO) and sectioned into five subsamples in 10-cm increments. Samples from each plot were then composited by depth, allowed to air dry, then ground to pass a 2-mm sieve, and analyzed for pH, EC, OM, and various macro- and micronutrients, at the Colorado State University Soil, Water, and Plant Testing Laboratory.
Soil pH and EC were analyzed using a saturated paste extract based on the methods of Miller and Kotuby-Amacher (1995). Soil organic matter was determined based on the methods of Self and Rodriguez (1997) by reacting SOM with K2Cr2O7 and H2SO4. To determine soil NO3–N, plant available P and other nutrients, soil samples were extracted using 20 mL of ammonium bicarbonate-DTPA (diethylenetriaminepentaacetic acid) (AB-DTPA extractant) and 10 g of soil (Self and Rodriguez, 1997). The extracts were analyzed for NO3–N content using flow-injection Cd reduction analysis, and extractable P was measured at 882 nm on a Spec 20 at an acidity of 0.18 M H2SO4 by reacting a sample aliquot with ammonium molybdate using ascorbic acid as a reductant in the presence of antimony. The AB-DTPA extracts were analyzed for Ca, Mg, K, Zn, Fe, Mn, and Cu by inductively coupled plasma spectroscopy (ICP).
Water Quality
The soil samples described above for soil chemical analysis were also used to evaluate NO3–N and P concentrations below the rootzone. The 30- to 40- and 40- to 50-cm depths were chosen to represent the region below the rootzone based on other research in the same plots (Johnson et al., 2006).
Rainfall simulations were conducted on the turf plots from 27 May through 4 June 2004. Plots had a 1 to 2% slope. A custom-made steel frame (1.27 x 2.03 x 0.15 m, W x L x H) was driven 10- to 13-cm deep into the undisturbed soil directly in front of the plot, snuggly against the edge of the turf. The top of the trough was set approximately at the soil surface.
The simulator (Joern Inc., West Lafayette, IN) was constructed of aluminum piping, based on the design of Miller (1987). A hanging TeeJet HH-SS50WSQ nozzle (TeeJet Agricultural Spray Products, Wheaton, IL) was used, which was approximately 3.048 m above the turf canopy. The simulator used a pressure regulator set at 69 kPa, which delivered a rainfall intensity of 6.7 cm h–1. One rainfall simulation was conducted on each of the 12 ‘Nuglade’ plots for 90 min each. Runoff was collected in 1-L polyethylene bottles for 2 1/2 min, in 5-min intervals during the first 30 min. After that, runoff was collected in 2 1/2 min intervals continuously for the remaining hour. Runoff samples collected during time intervals ending at 2.5 min and every 5 min thereafter up to 87.5 min were used to measure runoff and sediment loss. Sediment mass was determined following sedimentation and drying at 105 C for 24 h. Runoff and sediment loss rates were calculated by dividing the mass by the sampling interval (2.5 min). Runoff samples collected at time intervals ending at 35 min and every 5 min thereafter up to 90 min were used to measure nutrient content. Two 50-mL subsamples were taken from these samples. One was filtered (0.45-µm pore diameter), and the second was not filtered but was acidified to pH 2 with HCl. In a few cases, insufficient amounts of runoff were collected at certain time intervals, which allowed analysis for filtered samples only (see Statistical Analysis below).
An additional filtering process was required because the water used in the rainfall event was fairly saline (EC = 2.8 dS m–1), and it left behind white flakes in the sediment sample. Five hundred milliliters of deionized water were added to the sediment and samples were shaken, allowing the salts to dissolve. The liquid was then poured into a glass micro-fiber filter (Whatman GF/C), which was attached to a vacuum. The samples were dried again at 105°C for 24 h, and weighed after drying, and the true sediment mass was obtained.
Filtered (0.45 µm) samples were analyzed for OP using the ascorbic acid colorimetric method of Murphy and Riley (1962), and TDP using the ICP (Pierzynski, 2000). Ammonium N and NO3–N were determined using the methods for chemical analysis of water and wastes described by United States Environmental Protection Agency (1983). Total P and N were analyzed according to the American Public Health Association methods (Greenberg et al., 1992).
Statistical Analysis
Data on soil hydraulic conductivity, water retention, bulk density, and individual soil chemical properties at 0 to 10, 10 to 20, 20 to 30, 30 to 40, and 40 to 50 cm were subjected to analysis of variance using SAS Proc GLM (SAS Institute, 2002) to test effects of compost treatments on the soil properties. Means were separated using a protected LSD at P 
0.05.
Data (by time interval and means across time) from filtered and unfiltered runoff samples, flow rate, and erosion rate were subjected to analysis of variance using SAS Proc GLM (SAS Institute, 2002) to test effects of compost treatments on TN, TP, NH4–N, NO3–N, TDP, and OP. Missing data occurred twice in the filtered samples and 11 times in the unfiltered samples due to insufficient runoff volumes. Data collected on the first test plot (33 m3 ha–1, Replicate 3) were not included in the analysis of runoff volume or sediment due to improper trough installation resulting in very low runoff collection rates.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
|
There was no significant difference in Kfs among treatment levels at the culmination of this study (Table 2), but there is an indication that Kfs tended to increase with compost application rate. The natural spatial variability in Kfs apparently overshadowed the more subtle treatment effects, and eight measurements per treatment were apparently inadequate. The Kfs values were relatively high for a clayey soil when compared with other studies (Carsel and Parrish, 1988; Aggelides and Londra, 2000; and Miller et al., 2002). It appears that the soil still required more time to reach saturation although the plots were watered before data collection plots.
Other studies have shown significant increases in laboratory-measured saturated hydraulic conductivity (Ks) when incorporating compost into the soil. Celik et al. (2004) examined the effects that compost, manure, and fertilizer had on soil physical properties, and found that compost and manure amendment at 25 Mg ha–1 significantly increased Ks. In fact, at a depth of 0 to 15 cm, compost increased Ks by 65%, compared with the control. They believed this may have been related to soil porosity, in particular macroporosity, since soils with high macroporosity generally have higher Ks. In addition, Aggelides and Londra (2000) showed that Ks was increased 32.5, 53.0, and 95.2% in loamy soil and 55.3, 97.4, and 168.5% in clay soil for the rates 75, 150, and 300 m3 ha–1 of compost, respectively. The increase in soil porosity and hydraulic conductivity is also supported by Mathers and Stewart (1980). Although there was not a significant difference in Kfs in our experiment, if compost treatments were continued for a longer period of time, saturated hydraulic conductivity may become significant.
There was no significant difference in pH among treatments at any of the measured soil depths (Table 3). The EC of the compost was 32.8 dS m–1; however, the long-term use of high EC irrigation water (2.8 dS m–1) resulted in soil EC greater than 4 dS m–1 for all treatments. The soil EC was significantly different in the 0- to 10-cm depth only. The compost treatments 66 and 99 m3 ha–1 averaged 25% higher EC values than the control. Cuevas et al. (2000) reported similar increases in soil EC due to compost application.
|
There was no difference in NO3–N among treatments in any of the soil depths (Table 3). Nonetheless, all of the compost treatments had higher soil P and K in the 0- to 10-cm depth, compared with the control. Soil P content of all compost treatments was 187% higher than the control. Soil K content of the 33 m3 ha–1 compost treatment was 75% higher than the control, and the 66 and 99 m3 ha–1 compost treatments averaged soil K concentrations 146% greater than the control. Compost is a source of P and K, and as compost rates increased, so did the soil concentrations of each element. These results are supported by other studies (Schlegel, 1992; Cuevas et al., 2000).
There was no difference in Zn concentrations except for at the final depth of 40 to 50 cm (Table 3). Since only the 66 m3 ha–1 treatment was higher than the control at this one depth, it is believed that random soil variability and laboratory error contributed to this difference. However, compost treatments did play a role in Fe and Mn concentration in the 0- to 10-cm depth. The 66 and 99 m3 ha–1 compost treatments averaged 15.6% higher soil Fe concentration and 35.3% higher soil Mn concentration than the control. The additional Fe provided by the compost could be beneficial to the turfgrass, especially in helping to prevent Fe chlorosis in high pH soils. There were no significant differences in Cu concentrations at any soil depth.
Water Quality
Nitrate-N concentrations below the rootzone (30–50 cm) were not different among treatments (Table 3). Guillard and Kopp (2004) evaluated topdressing turf with turkey litter compost in a humid environment and found that <1% of N applied in this form leached annually, considerably less than nitrate leaching from ammonium nitrate fertilizer (17%). However, in this study, a single soil sampling cannot be extrapolated to predict nitrate leaching flux.
Treatments that received compost were significantly higher in AB-DTPA extractable soil P compared to the control in the surface depth (Table 3). Based on Colorado State University's fertilizer recommendations (Mortvedt et al., 2004), compost treatments resulted in medium (4–7 mg kg–1 P) concentrations, and the control was in the low (0–3 mg kg–1 P) range in the surface depth. However, there were no significant differences among treatments from 10- to 50-cm deep, and the concentrations decreased with depth. Therefore, neither soil NO3–N nor P concentrations below the rootzone were affected by topdressing compost on turf up to the 99 m3 ha–1 application rate.
In the rainfall simulation study, there were no significant differences among treatments in the runoff and erosion rates (Fig. 2 ). Runoff and erosion rates tended to increase with time and did not appear to reach a steady state by the end of the simulation (90 min). Under this worst-case scenario simulation (very high rainfall intensity <1 mo after the third compost application), no difference in TN or TP concentrations was observed among compost application rates (Table 4). The TP concentration was higher than the EPA standard of 0.1 mg L–1 for rivers (USGS, 1995) for all treatments, and most of it was contributed from the sediment (average was > 94% in all treatments). However, experimental error occurred due to additional sediment entering the collection bottles at the point where the trough met the soil. This contributed to higher amounts of sediment collected and has probably led to overestimation of TP. In addition, small plot runoff often overestimates edge-of-field losses due to the occurrence of soil redeposition in larger areas.
|
|
Extractable soil P has often been positively correlated with dissolved P in runoff (Pote et al., 1996). However, in this study, although extractable soil P was increased by compost application, there were no significant differences detected in TDP or OP by time increment or in the overall means (Table 4). There was no evidence that topdressing composted manure on turf led to any additional P hazards in runoff. This may be due to the similarities between turf and grass buffer strips mentioned earlier.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
As hypothesized, compost increased AB-DTPA extractable P, K, Fe, and Mn concentrations in the 0- to 10-cm depth. Compost applications also raised the EC of the soil in the surface depth, as expected. However, the high EC of the irrigation water resulted in all treatments having soil EC greater than 4 dS m–1. Even though the EC was high, there was no noticeable impact on the turfgrass (Johnson et al., 2006). The OM content was significantly increased in the 10- to 20-cm depth only. Other studies have shown that incorporating compost into the soil increases the OM content, as well (Schlegel, 1992; Aggelides and Londra, 2000; Egashira et al., 2003).
There were no differences in the amounts of soil NO3–N or P below the rootzone. Based on these results on a uniform 1 to 2% slope, topdressing compost onto turf resulted in no measurable difference in soil NO3–N levels or P runoff in 2 yr of application. The data is limited to one rainfall simulation, which although it does represent a worst-case scenario, cannot be used to predict a flux over the entire 2-yr period. In addition, a single soil sampling at the end of the study period may not capture nitrate leaching events during that 2-yr period. The soil is a clay loam located in a semiarid area (annual precipitation was 221 and 310 mm in the study years) with high evapotranspiration demand (1024 and 933 mm yr–1). However, irrigation may still result in some leaching if applied in excess of water requirements.
Therefore, topdressing composted cattle manure onto turf grown on clay loam soils can improve urban and suburban soil quality without degrading water quality under semiarid conditions. Results from this 2-yr study should not be extrapolated to conclude that long-term application of compost to turf at similar rates would have no adverse impacts. These application rates are recommended for soil quality improvement not for continued long-term use. To avoid long-term water quality impacts from continued compost applications, it will be of critical importance for turf managers to soil sample regularly and base compost use on soil test results and/or plant nutrient requirements, as appropriate based on local recommendations. In addition to the benefits to turfgrass, pressure on water quality may be reduced by transporting manure away from concentrated animal feeding areas to urban and suburban turf sites.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 1, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
Journal of Environmental Quality |
||||
