JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 26 April 2006
Published in J Environ Qual 35:903-911 (2006)
DOI: 10.2134/jeq2005.0266
© 2006 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Agricola
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Related Collections
Right arrow Animal Waste
Right arrow Runoff
Right arrow Water Pollution
Right arrow Nitrogen
Right arrow Phosphorus

TECHNICAL REPORTS

Waste Management

Aerating Grassland before Manure Application Reduces Runoff Nutrient Loads in a High Rainfall Environment

L. J. P. van Vlieta, S. Bittmana,*, G. Derksenb and C. G. Kowalenkoa

a Pacific Agri-Food Research Centre, Agriculture and Agri-Food Canada, P.O. Box 1000, Agassiz, BC, Canada V0M 1A0
b Environmental Protection Branch, Environment Canada, 224 West Esplanade, North Vancouver, BC, Canada V7M 3H7

* Corresponding author (bittmans{at}agr.gc.ca)

Received for publication July 7, 2005.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The effect of mechanically aerating grassland before liquid manure application in the fall on surface runoff and transport of nutrients and solids was studied in a high rainfall area. The two treatments were control and aeration, the latter receiving one pass with an aerator perpendicular to the slope before fall application of liquid manure (dairy in Years 1–3 and swine in Year 4). Treatments were randomly assigned on 3 to 5% sloping land with a silt loam surface soil (Aquic Dystroxerept) planted in orchardgrass (Dactylis glomerata L.). Runoff from natural rainfall events was sampled for nutrient and solids analysis. Aeration significantly reduced runoff and loads of suspended solids, total Kjeldahl N (TKN), and dissolved reactive P in all years. Annual runoff amounts were reduced by 47 to 81%, suspended and volatile solid loads by 48 to 69% and 42 to 83%, respectively, TKN loads by 56 to 81%, and total P (TP) loads by 25 to 75%. Loads of the soluble nutrient NH4–N, dissolved reactive P, and K were reduced by 41 to 83%. The first three runoff events after manure application accounted for approximately one-third of the annual total runoff and solid and nutrient loads when averaged across treatments, with loads of TKN, K, and NH4–N totaling 4.4, 3.3, and 1.9 kg ha–1, respectively. Aeration slightly increased downward movement of NO3–N, but not other nutrients in the soil. Thus mechanical aeration can be an effective tool for reducing runoff and loads of solids and nutrients after surface application of liquid manure on sloping grassland.

Abbreviations: DRP, dissolved reactive phosphorous • EC, electrical conductivity • SS, suspended solids • TKN, total Kjeldahl nitrogen • TP, total phosphorous • VS, volatile solids


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
LIVESTOCK PRODUCTION may impact surface and ground water quality (Sharpley et al., 1998). In intensively farmed areas in Canada, where large quantities of manure and fertilizer are used, concentrations of nutrients often exceed one or more water quality guidelines for surface water (Chambers et al., 2000). The impact of agricultural activities on water quality is evident in the intensively farmed Lower Fraser Valley of British Columbia (Van Vliet et al., 2002; Top et al., 1997; Servizi and Gordon, 1990). Manure application on grassland is susceptible to loss in surface runoff, depending on slope and other soil characteristics, because the manure is usually surface applied (Bittman et al., 1999). Injection into perennial grass swards is rarely done because of damage to the sward or difficult soil attributes such as stones. Runoff transport from surface-applied manure depends on application rates, rainfall intensity, and site characteristics such as slope and surface compaction.

Mechanical soil aeration in grassland has been suggested as a management practice to relieve surface compaction and improve forage stand health (Bertrand et al., 1991). A typical soil aerator has angled rolling tines that create surface slots of variable depth and length. Mechanical aeration of grassland has been used in England as a means to reduce soil bulk density from compaction (Douglas et al., 1995; Davies et al., 1989). In Canada, Gordon et al. (2000) used mechanical soil aeration immediately before and after spreading liquid dairy manure in Nova Scotia, but found no beneficial effects on NH3 volatilization and forage yields. More recent studies on perennial grass in British Columbia showed that surface-banding slurry (40-mm-wide bands spaced 23 cm apart) directly over aeration slots as a single operation, using the new SSD applicator (Holland Equipment Ltd., Norwich, ON), reduced emission of NH3 and odor and increased yield compared with surface broadcasting (Bittman et al., 2005; Lau et al., 2003). Crawford and Douglas (1993) and Douglas et al. (1995) speculated that mechanical soil aeration could also be used as a means to reduce runoff from grassland by improving water infiltration, but no data were presented. A recent study on grass in Arkansas using simulated rainfall showed that runoff of nutrients from poultry litter was reduced when the litter was placed into aeration slots made across a slope, particularly in the first runoff event after manure application (Pote et al., 2003). In a microplot study with simulated rainfall in West Virginia, mechanical aeration reduced nutrient runoff loss from dairy slurry that was broadcast on a perennial grass sward, but the effect was limited to manure applications directly after aeration (Shah et al., 2004). Neither study examined the effect of aeration on movement of nutrients through the soil profile. Our objective was to examine the effect of mechanical aeration of a grass sward before broadcast or band application of liquid manures on losses of solids and soluble nutrients (N, P, and K) by runoff and leaching under high-rainfall winter conditions.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Experimental Site
The 4-yr study was conducted near Agassiz, BC (49°16' N, 121°45' W) on a site with an irregular slope of 3 to 5%. The alluvial soil is imperfectly drained and belongs to the Fairfield series (Aquic Dystroxerept; Luttmerding, 1981). The surface soil consists of up to 25 cm of silt loam to very fine sandy loam and is underlain by fine sandy loam to loamy sand materials with a massive structure (van Vliet et al., 2002). The site, previously in silage corn (Zea mays L.), was prepared on 11 May 1998, by plowing, disking, cultivating and packing, then broadcast seeded the next day with orchardgrass (cv. Profile) at a rate of 25 kg ha–1. Ammonium nitrate fertilizer was applied (30 kg N ha–1) at seeding. The plots were sprayed (29 May 1998) with a tank mix of dicamba (3,6-dichloro-2-methoxybenzoic acid at 0.3 L a.i. ha–1) and 2,4-D [(2,4-dichlorophenoxy)acetic acid at 0.8 L a.i. ha–1] to control broadleaf weeds. The crop was harvested once in the seeding year and three to five times in subsequent years before initiating trials in autumn.

Each runoff plot was 21 by 6 m with the long side oriented downslope. The plots were isolated on three sides by metal edging, set 5 to 7.5 cm into the ground, and a covered gutter (10 cm wide and 8 cm deep) to collect the runoff across the downslope width of the plot. The gutter, installed flush with the soil surface, drained by gravity into a 900-L collecting tank connected to a 2800-L reserve tank to collect overflow, both buried in the soil. The two tanks, combined, stored up to ~30 mm of runoff from the plot area. A pair of porous ceramic suction cup lysimeters was installed (60- and 120-cm depths) at both the upper and lower parts of each plot.

Before assigning treatments, plots were found to be similar in particle size, organic matter content, total N, pH, and bulk density (6-cm depth using 2.0-cm by 4.75-cm-diameter cores). Soil characteristics are shown in Table 1. The two treatments, mechanical soil aeration and an unaerated control, were randomized within each of three blocks, with the randomization changing each year. Aeration treatment was performed in late October, immediately after grass harvest. Aeration was performed across the slope with a 3.1-m-wide AerWay (100Q) soil aerator (Holland Equipment Ltd., Norwich, ON) set at 2.5° offset and traveling at 4 km h–1. The operation created 11 tapered slots m–2 in rows spaced 19 cm apart; slot dimensions were measured for 16 randomly selected slots per aerated plot immediately after aeration and at 2- to 3-wk intervals (Table 2). In Years 1 through 3, dairy slurry was broadcast the day after aeration. In Year 4, swine manure slurry was band applied over aeration slots in the same pass as the aeration using an Aerway SSD manure applicator (Holland Equipment Ltd., Norwich, ON), described in Bittman et al. (2005), while the control treatment was applied by broadcasting. The dates and rates of manure application and manure characteristics are shown in Table 3. Application rates targeted 100 kg ha–1 of total NH4–N with application volumes not exceeding the practical limit of ~9 m3. Total N in manure samples was measured on dried samples by colorimetry after Kjeldhal digestion using H2SO4, K2SO4, CuSO4·5H2O, and Se. Phosphorous and K were measured by inductively coupled argon plasma spectrophotometry on a HNO3 and HClO4 digest. Ammonium was determined by steam distillation in the presence of Mg.


View this table:
[in this window]
[in a new window]
  		Table 1. Soil properties at the runoff study site at Agassiz, BC.

 

View this table:
[in this window]
[in a new window]
  		Table 2. Dimensions of slots immediately after aeration and over the winter seasons (1998–2002).

 

View this table:
[in this window]
[in a new window]
  		Table 3. Dates and rates of manure application and the characteristics of dairy (1998–2000) and swine (2001) manure slurry used for aeration runoff study in south-coastal British Columbia.

 
Daily precipitation data were collected with a Universal Recording Rain Gauge (Belfort Instrument Co., Baltimore, MD) located at the site while historic precipitation data were obtained from a weather station located 3 km away (Fig. 1).


Figure 1
View larger version (25K):
[in this window]
[in a new window]
  		Fig. 1. Monthly precipitation at Agassiz during the study period and the long-term (1961–1990) average.

 
Runoff and Lysimeter Measurements
After the fall manure application, runoff tanks were checked at least once weekly. When accumulated runoff exceeded 1 L per plot, it was considered a runoff event and sampled for nutrient analysis. Runoff volume was determined by depth of water in the tanks. For volumes <25 L (3-cm depth in tank), samples were collected directly from the tank. For volumes >25 L, runoff water in the tanks was mixed with a pump while three depth-integrated samples (0.5 L) were taken (Brakensiek et al., 1979). When the primary tank overflowed into the secondary tank (six runoff events in Year 1), the entire volume from both tanks was combined before sampling. Tanks were emptied and carefully cleaned with a vacuum between runoff samplings. A sample (1 L) was frozen for analysis of concentrations of suspended solids (SS), NO3–N, NH4–N, and TKN. Another 1-L sample was refrigerated at 4°C for determination of pH and concentration of volatile solids (VS). A third sample (250 mL) for analysis of TP, dissolved reactive P (DRP), and total K was acidified to pH < 2 by adding 2 mL of 8 M HNO3, then refrigerated at 4°C. Lysimeter samples were collected weekly after manure application and analyzed for NO3–N, NH4–N, DRP, electrical conductivity (EC), and pH, as described below.

Suspended solids were determined gravimetrically after filtering (Brakensiek et al., 1979). The filtrate was analyzed for NO3–N and NH4–N with a flow injection analyzer (Tecator model no. 5010, Perstorp Analytical Co., Sweden) (Kowalenko and Yu, 1996). Total Kjeldahl N was determined on unfiltered samples after evaporating at room temperature and digesting with 18 M H2SO4; NH4–N was determined by steam distillation with a Kjeltec automated steam distillation–titration instrument (Tecator model no. 1030, Perstorp Analytical Co., Sweden). Concentration of VS, which is an approximation for the total amount of organic matter in the solid fraction, was determined gravimetrically by loss-on-ignition at 550°C (Eaton et al., 1995). Concentrations of TP and K were determined by inductively coupled plasma on samples digested with 18 M H2SO4. Dissolved reactive P in runoff water was determined on a filtered sample using the colorimetric method of Murphy and Riley (1962) as outlined by Pote and Daniel (2000). Sample pH and EC were measured with a glass electrode, using a Radiometer ION 83 pH/ion meter (Yellow Springs Instruments, Yellow Springs, OH). Nutrient loading for each runoff event was calculated as nutrient concentration x runoff volume.

Data Analysis
All data were analyzed with the GLM option in SYSTAT (Wilkinson, 1997) with significance level set a priori at P < 0.05. For the across-years analysis, a split-plot model with year as the main effect was used and, since year was significant, the data were analyzed by year. Results from individual years were analyzed using a randomized complete block design. Homogeneity of variances was ascertained using Bartlett's test so that pooled variances could be used for the error term.


   RESULTS AND DISCUSSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Based on the surface area of each slot (25 by 2 cm) and the number of slots per area (11 m–2), slot surface area immediately after treatment accounted for ~5% of the surface soil (Table 2). Assuming an average depth of 15 cm and a triangular-shaped slot, slot volume was estimated to be 375 cm3, or approximately 40 m3 ha–1 retention capacity, which was enough volume to contain more than half of the manure applied in the fall (Table 3). Initial slot depth can be set with depth wheels but depends also on the moisture conditions, texture, stoniness, and degree of compaction of the surface soil. Soil was moist to wet during aeration in three of the 4 yr (cumulative rainfall during 7 d before aeration: 58.0 mm in Year 1, 32.0 mm in Year 2, 8.1 mm in Year 3, and 69.0 mm in Year 4). While the aeration slots remained visible and functional throughout the period of runoff measurements, manure, sediments, and residues reduced their length and depth with time, particularly in November or December, thus reducing their effectiveness in intercepting surface runoff (Table 2). Herbage measurements taken in Years 2 and 3 showed that aeration treatments in the previous autumns did not influence annual dry matter yields or total N concentrations of the herbage (data not shown).

Winter-season (October–April) precipitation for Years 1, 2, 3, and 4 was 30% higher, 3% higher, 34% lower, and 11% higher, respectively, than the long-term (1961–1990) average of 1156 mm (Fig. 1). The wettest period during the study was November to December 1998, with a cumulative precipitation of 804 mm (41% of the annual precipitation), resulting in seven runoff events. The driest winter period during this study lasted from October 2000 to February 2001, with approximately half the long-term average precipitation. Snowfall was <10% of the winter-season precipitation in each year, which is similar to the long-term average (7.6%); there were up to three runoff events that contained snowmelt in each year of the study. Of the 56 winter runoff events recorded during the 4 yr of study, 16 runoff collections were made in Year 1, 15 in Year 2, 10 in Year 3, and 15 in Year 4. Almost half of the runoff events sampled in this study occurred in November and December. By far the largest runoff event (32 mm) in this study occurred in December 1998 and the second largest was in December 2001 (8 mm; Fig. 2). Cumulative runoff for the sampling period each year followed the same ranking (Year 1 > Year 4 > Year 2 > Year 3) as winter precipitation, with Year 3 runoff only ~6% that of Year 1.


Figure 2
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Runoff volume during runoff events following manure application on aerated and unaerated (control) sloping grassland during four winter seasons in coastal British Columbia. Arrows indicate dates of manure application.

 
Aeration significantly reduced the cumulative winter runoff compared with the control in all years (Table 4). In Year 1, runoff collection peaked on 29 Dec. 1998 after a 5-d period with 136 mm of precipitation. The aeration treatment reduced runoff volume during this period from 309 m3 ha–1 (31 mm) to just over 79 m3 ha–1 (7.9 mm; Fig. 2). Runoff in Year 2 showed three distinct but smaller peaks during November and December, with aeration reducing runoff from 18.6 m3 ha–1 (1.9 mm) to 8.7 m3 ha–1 (0.8 mm) during the highest of the peaks on 20 Dec. 1999. While runoff in Year 3 followed a similar temporal pattern as in Year 1, the volume was very small due to below-average precipitation. The largest seasonal reduction (81%) was in Year 1 when the newly seeded sward was still forming (Table 4). Similarly, mechanical aeration of pastures in Arkansas reduced the amount of surface runoff from simulated rainfall by 45% (Moore et al., 2003). The aeration slots probably help to reduce runoff first by directly accommodating up to 4 mm of rainwater. Further, we observed that the aeration process increases surface roughness, which would impede runoff, particularly when the slots are oriented across the slope.


View this table:
[in this window]
[in a new window]
  		Table 4. Effect of grassland aeration before application of liquid dairy manure (1998–2001) or liquid swine manure (2001–2002) on cumulative winter runoff and nutrient loads.

 
Another benefit of mechanical aeration is alleviating soil compaction by the soil shattering action between the rolling tines (Davies et al., 1989; Crawford and Douglas, 1993; Douglas et al., 1995). Shah et al. (2004) also reported that mechanical aeration reduced soil impediment and improved root penetration in compacted soils of grasslands in West Virginia. We measured no significant reduction in average bulk densities (6-cm depth) during the summer from aerations performed the previous autumn. The amount of runoff from sloping grassland in our study was much lower than often reported for cropland, where raindrop impact can seal the surface of exposed soil, reducing infiltration (Chow et al., 1990; van Vliet and Hall, 1991, 1995; Wall et al., 1991; van Vliet et al., 1997, 2002).

Concentrations of SS and VS in runoff events were generally highest when runoff amounts were low (not shown). Higher concentrations of sediment are usually expected with greater runoff volumes but this may not apply to very low flow conditions over manured grass stands, where dilution may be a more important factor. In Year 1, aeration increased concentrations of SS and VS in 10 of the 16 runoff events but the effect was pronounced only for the first runoff event when aeration increased SS concentration fourfold and VS concentration almost threefold (not shown). Aeration appeared to cause more disturbance of the soil surface in the first-year grass stand than in subsequent years when the sod was more fully formed. It is likely that this exposed and loosened soil contributed to the high SS concentrations. Shah et al. (2004) previously reported that mechanical aeration on grassland could significantly increase total SS concentrations in runoff, although, in our study, aeration performed in Years 2 through 4 when the orchardgrass sward was well established did not increase concentrations of SS and VS (not shown) irrespective of differences in seasonal rainfall in those years (Fig. 1). Concentrations of VS, like those of SS, were typically high during early runoff events in November and December and much lower during runoff events January to April (not shown).

The amount of runoff was a major factor determining nutrient and sediment loads, so load data generally followed a pattern that was similar to runoff volume. Cumulative SS and VS loads varied widely across years, with the two wettest years, Years 1 and 4, having the highest of both runoff and SS loads (Table 4); however, the SS load in Year 4 was much higher than in Year 1 (Table 4) because SS concentrations in Year 4 were many times higher than those in Year 1 (not shown). Temporal patterns for SS load (Fig. 3) and VS load (not shown) were characterized by one or more peak loads (up to 120 kg ha–1 for SS and 47 kg ha–1 for VS) in November and December, followed by much lower loads (<20 kg ha–1 for SS and <6 kg ha–1 for VS) from January through April.


Figure 3
View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Suspended solid loads for runoff events following manure application on aerated and unaerated (control) sloping grassland during four winter seasons in coastal British Columbia. Arrows indicate dates of manure application.

 
Aeration significantly lowered total SS loads relative to the control in each of the 4 yr (Table 4). Aeration also significantly lowered VS loads in Years 1 and 2. It is interesting that aeration was effective in reducing solid loading in wet and dry years, showing that the technique is likely to work under a wide range of conditions. As with runoff volume, SS loadings from grassland after manure application (<0.5 Mg ha–1 yr–1) in this study were much lower than those from cropland (7–10 Mg ha–1 yr–1) in previous studies conducted in the Lower Fraser Valley of British Columbia (van Vliet et al., 1997, 2002). In low-rainfall areas, runoff transported up to 1 Mg ha–1 of solids annually from fescue (Festuca spp.) and alfalfa (Medicago spp.) fields with comparable slopes (van Vliet and Hall, 1991; Hargrave and Shaykewich, 1997). A comparatively low quantity of total solids (7.9 kg ha–1) was removed by surface runoff from irrigated tall fescue (Festuca arundinacea Schreb.) treated with liquid dairy manure in South Carolina (McLeod and Hegg, 1984).

The seasonal pattern of loads of TKN was similar to that for runoff (Fig. 4) while seasonal TP loads (except for Year 4) also followed a similar temporal distribution (not shown). In spite of high TKN and TP concentrations in Year 3, the very low amount of runoff resulted in low cumulative TKN and TP loads. Aeration significantly lowered seasonal TKN loads compared with the control in each of the 4 yr and TP loads in 2 of the 4 yr (Table 4). For control plots, seasonal loadings of TKN relative to the amount of total N applied in the manure were 1.8% in Year 1, 0.6% in Year 2, and 0.2% in Year 3, and 5.6% in Year 4. For aerated plots, seasonal loadings of TKN were 0.36, 0.28, 0.06, and 0.84% of applied total N in Years 1 to 4, respectively. Likewise, for control plots, seasonal loadings of TP relative to the amount of total P applied in the manure were 4.3% in Year 1, 0.3% in Year 2, 0.1% in Year 3, and 0.8% in Year 4. For aerated plots, seasonal loadings of TP were 3.21, 0.07, and 0.02% of applied total P for Years 1 to 3, respectively (Year 4 data are not available). Averaged across the study periods, aeration reduced TKN and TP loads by 71 and 83%, respectively (Table 4). In a runoff study using simulated rainfall, Moore et al. (2003) found that aeration of grassland reduced the load of total N by 55% and total P by 43%. The annual TKN and TP loads in our study are similar to the 2.3 kg TKN ha–1 and 1.8 kg TP ha–1 reported by McLeod and Hegg (1984). Annual loads in runoff from grassland treated with dairy slurry under natural rainfall in Finland varied from 0.7 to 8.0 kg ha–1 for total N and from 0.2 to 2.7 kg ha–1 for TP (Uusi-Kamppa and Heinonen-Tanski, 2001).


Figure 4
View larger version (28K):
[in this window]
[in a new window]
  		Fig. 4. Total Kjeldahl nitrogen (TKN) loads for runoff events following manure application on aerated and unaerated (control) sloping grassland during four winter seasons in coastal British Columbia. Arrows indicate dates of manure application.

 
Other nutrients measured in the runoff—NO3–N, NH4–N, DRP, and K—are soluble and, therefore, their temporal distributions are less directly related to sediment concentrations than TKN and TP. Except for the concentration peaks of NH4–N, DRP, and K on control plots during the first annual runoff events each year, no distinct temporal patterns were observed in concentrations of the soluble nutrients (not shown). Noteworthy is the high NH4–N concentration peak (102 mg L–1) for the unaerated treatment during the first runoff event in Year 4 compared with both the aerated treatment in the same event (7.3 mg L–1) and other control treatment values (1–15 mg L–1) when dairy manure was applied.

There were one or more peaks per year in soluble nutrient loads for the control treatment. For example, NH4–N loads peaked at every first annual runoff event, with the highest load (0.94 kg ha–1) during the 4-yr study occurring in November 2001 (Fig. 5). The January through April periods were always characterized by very low loads (<0.01 kg ha–1 of NO3–N, NH4–N, and DRP). For the control treatment, maximum annual loads of soluble nutrients were observed in Year 1 for K (6.0 kg ha–1 or 5.4% of applied K) and in Year 4 for NH4–N (2.2 kg ha–1 or 2.4% of applied NH4–N). Nitrate-N and DRP had annual loads of <1 kg ha–1 (Table 4), similar to the loads reported by McLeod and Hegg (1984) and Uusi-Kamppa and Heinonen-Tanski (2001). Except for K, all annual soluble nutrient loads for the aerated treatment were <0.75 kg ha–1, with a minimum annual load of 0.01 kg ha–1 for NO3–N in Year 3. Aeration reduced the loads of DRP in each of the 4 yr but significantly reduced K and NH4–N loads only in Years 1 and 4 while loads of NO3–N were decreased in Year 1 and increased in Year 3 (Table 4). Pote et al. (2003) reported that, for grassland treated with poultry litter, aerated plots tended to have better runoff quality after simulated rainfall than unaerated plots, but the effect of aeration was not consistent for all runoff events. Shah et al. (2004) reported that mechanical aeration of grassland before application of liquid dairy manure reduced total loadings of all nutrients (NO3–N, NH4–N, TKN, DRP, and TP) in runoff by at least 26%. Annual overwinter runoff losses of both TKN and TP were 0.2 to 4.5% of the amounts applied in the manure in the fall (Tables 3 and 4). Edwards and Daniel (1994) reported similar results for fescue grass plots treated with poultry litter, where total losses of N and P were 1.4 and 2.7%, respectively, of the amounts applied.


Figure 5
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 5. Ammonium-N loads for runoff events following manure application on aerated and unaerated (control) sloping grassland during four winter seasons in coastal British Columbia. Arrows indicate date of manure application.

 
In our study, the first three runoff events contributed one-third of the cumulative annual total of runoff volume and loads of VS and nutrients, and more than half of the DRP loads (Table 5). The proportion was even higher in Year 3 because the winter season turned out to be quite dry. Similar trends have been reported by other researchers. In Kentucky, ~50% of the pollutant load from manured grassland occurred in the first 20% of total runoff (Ross et al., 1979). In South Carolina, concentrations of potential pollutants from surface-applied nutrients to a fescue pasture were reduced between 55 and 80% after two runoff events (McLeod and Hegg, 1984). Also, the first three runoff events after fall manure application accounted for >50% of the annual solid and nutrient loads from silage corn land (van Vliet et al., 2002). A short time interval between manure application and first runoff increases the risk of surface water contamination (Ross et al., 1979; Bottom et al., 1983; Edwards et al., 1996). In our study, the first runoff event occurred 25, 5, 31, and 1 d after manure application in Years 1 through 4, respectively. In Year 4, manure was applied on relatively wet soil after a week with 69 mm of rainfall and just before 24 mm of rain, which resulted in a runoff event the day after application. This helps to explain the high loads observed in this runoff: 4.4, 3.3, and 1.9 kg ha–1 for TKN, K, and NH4–N, respectively. The 5-d interval between manure application and first runoff combined with lower runoff volumes may explain the lower peak values in nutrient loads in Year 2.


View this table:
[in this window]
[in a new window]
  		Table 5. Volume of runoff and loading of solids and nutrients (average across treatments) in the first three runoff events as proportion of winter totals from a sloped grassland site receiving liquid manures in south coastal British Columbia (1998–2002).

 
Concentration of TP in most runoff events exceeded British Columbia water quality criteria for aquatic life (0.005–0.015 mg L–1), whereas the NO3–N concentration fell within the guidelines (maximum of 200 mg L–1, and 30-d average of <40 mg L–1) (British Columbia Environment, 1998). When considering pH and temperature, the maximum allowable NH4–N concentration of 20 mg L–1 designed to protect freshwater life (British Columbia Environment, 1998) was exceeded only during the first two runoff events in Year 4. The concentration of 102 mg NH4–N L–1 in the first of these events is rated as acutely toxic to fish (Servizi and Gordon, 1990). Nitrate loads in this study were small and probably of limited concern for surface water quality. There are no water quality criteria for K in British Columbia. All concentration parameters measured in this study fell within the Canadian drinking water quality criteria (Health and Welfare Canada, 1993).

Data for lysimeter samples are summarized in Table 6. There was little indication that aeration consistently increased downward movement of nutrients in the soil profile. No significant increases were found for NH4–N or PO4–P, although a rather large but nonsignificant increase due to aeration was observed for PO4–P in Year 1. There was also no consistent effect on EC or pH. The highest NO3–N concentrations for both treatments were observed in Year 3, which had the lowest precipitation. The high concentrations in Year 3 were associated with low runoff from the plots and suggest that low concentrations in other years are probably associated with dilution. Also, low concentrations of NO3–N in autumns and winters previously reported from nearby manured fields may also be due to dilution (Patni et al., 2000). It would be useful to compare the amount of NO3–N and PO4–P lost by leaching under aeration and nonaeration in relatively wet vs. dry winters.


View this table:
[in this window]
[in a new window]
  		Table 6. Effect of soil aeration before application of liquid manures on average quality of soil solution extracted weekly during the winter with suction lysimeters from two soil depths (1998–2002) in a runoff trial in coastal British Columbia.

 
The two manure application methods, broadcasting in Years 1 through 3 and banding over the aeration slots in Year 4, could not be compared in this study because of contrasting rainfall each year and because different manures were used (dairy in Years 1–3 and swine in Year 4). While methods could not be compared, it is apparent from the results that aeration was effective across a wide range of manure characteristics and weather conditions and that aeration could be used in conjunction with both surface broadcasting and banding. Surface banding may be preferred because it reduces odor (Lau et al., 2003) and gaseous NH3 loss (Bittman et al., 2005), whereas broadcasting over aeration slots may fail to reduce NH3 emission (Gordon et al., 2000). Losses of NH4–N by runoff in this trial were much lower than emission values reported by Bittman et al. (2005).


   CONCLUSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Conducted under natural wintertime precipitation, our study showed that mechanical aeration before surface application of liquid manure can be an effective tool for reducing runoff volume and solid and nutrient loads from a sloping grassland. The beneficial effects of aeration were observed under greatly contrasting amounts of rainfall. During the study, aeration reduced total runoff volume by 47 to 81%, SS loads by 48 to 69%, VS loads by 42 to 83%, TKN loads by 56 to 81%, and TP loads by 25 to 75%. Aeration reduced loads of soluble NH4–N by 56 to 81% and DRP by 60 to 96%. The difference between manure application methods used in Years 1 through 3 (broadcast over the aeration slots) and Year 4 (banded over aeration slots) could not be inferred from this study because of differences in rainfall and type of manure used. Nevertheless, it was evident that aeration effectively reduced nutrient loss with both methods of manure application. Runoff events occurring immediately after manure application may contain high levels of nutrients, even when grassland has been aerated. Even though the aeration slots gradually filled and closed, and probably became less effective throughout the winter, they appeared to remain functional for several months. There was low risk of significant surface transport of nutrients after the first three runoff events in November and December, particularly with aeration. Aeration slightly increased leaching of NO3 in the soil, but there was little evidence of leaching of other nutrients. The results show that mechanical aeration produces little soil disturbance and supports the observation of little damage to grass plants (Bittman et al., 2005), in contrast to some injection tools (Rodhe and Etana, 2005). To minimize the risk of surface water contamination, manure should be applied only when there is low probability of substantial rain for several days and an unmanured buffer strip should be used. (It would be useful to determine if the buffer strip should be aerated.) Therefore, mechanical aeration of grassland appears to be a practical method for controlling runoff and associated nutrient transport from surface-applied manure, without reducing grass yield.


   ACKNOWLEDGMENTS
 
The authors wish to express their appreciation to E. Kenney, L. Birston, G. Wilson, C. Van Laerhoven, D. Babuin, M. Richtier, and C. Engel for their technical assistance. The financial support by Environment Canada through the Georgia Basin Ecosystems Initiative is gratefully acknowledged.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
PARC Contribution no. 724.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
D. R. Lapen, E. Topp, M. Edwards, L. Sabourin, W. Curnoe, N. Gottschall, P. Bolton, S. Rahman, B. Ball-Coelho, M. Payne, et al.
Effect of Liquid Municipal Biosolid Application Method on Tile and Ground Water Quality
J. Environ. Qual., May 1, 2008; 37(3): 925 - 936.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
D. M. Butler, D. H. Franklin, M. L. Cabrera, A. S. Tasistro, K. Xia, and L. T. West
Evaluating Aeration Techniques for Decreasing Phosphorus Export from Grasslands Receiving Manure
J. Environ. Qual., May 1, 2008; 37(3): 1279 - 1287.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Agricola
Right arrow Articles by van Vliet, L. J. P.
Right arrow Articles by Kowalenko, C. G.
Related Collections
Right arrow Animal Waste
Right arrow Runoff
Right arrow Water Pollution
Right arrow Nitrogen
Right arrow Phosphorus

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