JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

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 Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in 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 Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Agricola
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Related Collections
Right arrow Grazing Management
Right arrow Nitrogen
Right arrow Lysimeter/Rhizosphere Studies
Right arrow Plant and Environment Interactions
Right arrow Water Pollution
Published in J. Environ. Qual. 33:637-644 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORT

Plant and Environment Interactions

Nitrate Leaching under Grassland as Affected by Mineral Nitrogen Fertilization and Cattle Urine

M. L. Decau*, J. C. Simon and A. Jacquet

UMR-INRA950, Université de Caen, Esplanade de la paix, 14032 Caen, France

* Corresponding author (cliquet{at}ibfa.unicaen.fr).

Received for publication April 18, 2003.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
An experiment was performed to better understand to what extent nitrogen fertilization rate and date and amount of urine deposition, when acting in combination, influence nitrate leaching under grassland. Leaching was studied during two successive winters using 2-m2 grassed lysimeters under three levels of N fertilization (0, 150, and 300 kg N ha–1 yr–1, referred to as 0N, 150N, and 300N, respectively), two levels of 15N-labeled urine (105 and 165 kg N ha–1, referred to as A2 and A3, respectively), and three dates of urine application (spring, summer, and fall). During the first winter, total N leaching losses varied between 2 and 50 kg N ha–1. When tested in combination, N applied as urine to grassland resulted in three times the total N loss by leaching that occurred following N fertilization in the first winter (4.3, 20.8, 34.9, 14.2, 17.1, and 28.7 kg NO–3–N ha–1 for no urine, A2, A3, 0N, 150N, and 300N, respectively). Leaching of 15N urine significantly depended on the date of application: 6.6, 17.3, and 29.1 kg for spring, summer, and fall, respectively. A similar pattern was observed for the contribution of 15N urine to total N leaching with 4.3, 12.9, and 21.4%. However, urine application, both in terms of amount and date, showed very little long-term effect on these N losses in Year 2. In our conditions of low winter rainfall and drainage, grazing management (through season, urinary N amounts, and urine N concentration) resulted in a higher impact on water nitrate quality than moderate N fertilization management.

Abbreviations: 0N, nitrogen fertilization of 0 kg N ha–1 yr–1 • 150N, nitrogen fertilization of 150 kg N ha–1 yr–1 • 300N, nitrogen fertilization of 300 kg N ha–1 yr–1 • A2, treatment providing 10.5 g N m–2 as urine • A3, treatment providing 16.5 g N m–2 as urine


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
USE OF NITROGEN FERTILIZER is considered essential as a management tool for systems involving grazing livestock for intensive dairy production. A recent UK survey (Jarvis, 1999) showed that mean yearly N application rate on dairy farms swards amounted to 281 kg N ha–1 (varying from 100–689 kg N ha–1). Such high N fertilization rates allow for large grassland and forage production, sustaining high animal densities (Jordan and Smith, 1985; Jarvis et al., 1989).

Grazing animals generally use only 10 to 35% of the N ingested, with the remainder returning to the soil in urine and feces (Bussink, 1994). Excreted N is mainly returned in urine (Deenen and Lantinga, 1993). Animal excreta are a valuable source of N and other nutrients for plants (Silva et al., 1999). However, there has been growing concern about the contribution of excreted N to the losses of N from grazed grassland. Previous work has indicated that under grazed grassland, cattle urine is a major contributor to NO–3 water contamination (Hack-ten Broeke and van der Putten, 1997).

The total N content of urine varies between 2.5 and 13 g kg–1 according to the N content of the forage and any supplemental feed, animal type, and age (Deenen and Middelkoop, 1992; Peyraud et al., 1995). Application of 100 to 800 kg N ha–1 yr–1 on the area actually receiving urine can greatly exceed plant requirements. Thus, excess N may leach to the ground water, be lost to the atmosphere, or be immobilized in soil organic matter (Jarvis et al., 1989; Lockyer and Whitehead, 1990; Williams and Haynes, 1994). Nitrogen leaching losses have largely been assessed according to N fertilization rates, or else have been determined for urine patches in different conditions of soil and climate (Barraclough and Jarvis, 1989; Clough et al., 1998; Stout et al., 1998). Based largely on studies where only one factor was tested, it is well established that N fertilization rate and date and amount of urine deposition influence N leaching under grazed grassland, but the combined effects of these three factors have not been studied yet.

The objectives of this study were to (i) study and characterize different parameters of NO–3 leaching losses (drainage amount, nitrate content of drained water, and N leaching) as affected by cattle urine N content, season of urine deposition, and mineral N fertilization rate; and (ii) determine the relative impact and importance of each factor on the leaching losses, when studied in combination. The impact of these different factors was studied in a lysimeter experiment to allow accurate leaching measurements (Addiscott, 1990), but the experiment differed from the majority of lysimeter studies found in the literature, as only 20% of the lysimeter area was treated with urine thus representing what usually occurs on grazed grasslands (Richards and Wolton, 1976; Cuttle et al., 2001). Moreover, this experiment examined both the one- and two-year influence of urine deposition on N leaching under grassland.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The study was conducted at Le Robillard, France (48°59' N, 0°0' W). In 1993, 20 lysimeters were filled with 0.9 m of calcosol repacked per native soil horizons (Association Française pour l'Etude des Sols, 1995) from the site. Relevant properties are presented in Table 1. The bottom 0.1-m section of each lysimeter was filled with quartz gravel to ensure a free-draining situation. Volumetric soil water contents were measured in situ during the experiment using time domain reflectometry (TDR) (Trime.FM; IMKO Gmbh, Ettlingen, Germany). The water-holding capacity (121 mm for the 0.9-m soil depth) was calculated as the difference between the soil water content at field capacity and the minimum value of soil water content observed during the experiment. The cross-sectional area of each lysimeter was 2 m2.


View this table:
[in this window]
[in a new window]
  		Table 1. Typical properties{dagger} of the soil.

 
The lysimeters, buried in the soil so that their surface was at ground level, and the soil surrounding them were sown with 25 kg ha–1 of ryegrass (Lolium perenne L. cv. Magella) in spring 1994. The investigation was performed from 1996 to 1998. Until 1996, the whole area was mown five times a year down to a sward height of 4 cm, a typical value for grazing cattle, and top-dressed with mineral fertilizer at 15 g N m–2 yr–1.

Over a 10-d period, about 200 L of urine were collected (and immediately frozen) from a cow fed exclusively on silage maize (Zea mays L.), and kept in a metabolism crate. Before use, it was unfrozen, mixed, and enriched with a solution of 15N-labeled urea to obtain two urine lots (A2 and A3) differing only in total N, urea N, and 15N excess. Initial and final chemical characteristics of the urine lots are shown in Table 2.


View this table:
[in this window]
[in a new window]
  		Table 2. Chemical characteristics of the collected and applied urine lots.

 
During the 2-yr measurement period, the sward was cut five times per year at a 4-cm height and clippings removed to simulate cattle grazing on the following dates: 4 May 1996, 5 June 1996, 15 July 1996, 28 Aug. 1996, 23 Oct. 1996, 15 Apr. 1997, 13 May 1997, 18 June 1997, 22 July 1997, 15 Sept. 1997, and 30 Mar. 1998.

In 1996 and 1997, six lysimeters received 0 N fertilization (0N), seven lysimeters received 15 g N m–2 yr–1 (150N), and the last seven lysimeters received 30 g N m–2 yr–1 (300N) in five dressings per year of ammonium nitrate, before each period of regrowth. One lysimeter in each series did not receive urine and acted as control (0N, 150N, 300N). After cutting in spring (4 May 1996) one lysimeter from each N treatment received 3 L of 15N-labeled A2 urine and one other lysimeter in each N treatment received 3 L of 15N-labeled A3 urine on a 0.4-m2 area in the center of the lysimeter. Application of urine supplied either 52.5 g (A2) or 82.5 g (A3) N m–2 (equivalent to 10.5 g and 16.5 g N m–2 over the whole lysimeter, respectively). The urine applications (A2 and A3) were repeated on a different set of lysimeters from the three N treatments in summer (15 July 1996) and to a further set in the fall (23 Oct. 1996), except for A3 in summer on 0N lysimeter. We considered 3 L as an average amount of urine for dairy cows on the basis of a previous study by Farruggia and Simon (1994). On these three occasions, lysimeters that did not receive urine received instead a similar volume of water. Thus, the 20 lysimeters held the following treatments: controls for N fertilization (0N, 150N, 300N) and 17 treatments of three N rates x three application dates of urine x two N contents of urine, except 0N summer A3.

Potassium and phosphorus fertilizers were applied on 13 Mar. 1996 and again on 13 Mar. 1997 at 15 g K and 4 g P m–2, respectively, to ensure favorable grass growth conditions.

Annual climatic averages (1971–2000) of the site were 642 mm rainfall, which was evenly distributed, and 10.5°C mean annual air temperature. Daily weather information was recorded using an automatic meteorological station located 200 m from the plot.

Over the two experimental years, mean air temperature varied between 10°C in 1996 (Year 1) and 11°C in 1997 (Year 2). Rainfall pattern differed from the long-term recorded data. In the summer of 1996 (from June to August) a rainfall deficit of 83 mm developed. This represented only 40% of the usual precipitation in this location at this time of year. To prevent the swards from drying out, we applied 90 mm of irrigation water in three applications of 30 mm with a watering can. During the first 4 mo of 1997 (first winter) rainfall was also very small and again we applied 90 mm of irrigation water, as described above, to ensure that normal drainage would result from typical rainfall events. During the last winter (October to December 1997), though rainfall was as expected, we added 90 mm of irrigation to maximize soil leaching.

Drainage water was collected and the volume measured. The solution was analyzed weekly for NO–3–N and 15N using autoanalyzer colorimetric techniques (Bran+Luebbe GmbH, Norderstedt, Germany) and a Carlo Erba (Milan, Italy) instrument linked to an Isotope Ratio Mass Spectrometer (Europa Scientific, Crewe, UK).

Analysis of variance (ANOVA), as well as step-by-step regressions, were performed by Stat-ITCF (Institut Technique des Céréales et des Fourrages, 1987). In the step-by-step regression, the model sought the factor responsible for the largest proportion of the variance of the studied parameter, calculated r2, and then added a second factor that further explained the variance, then recalculated r2 with two factors and so on with the remaining tested factors. This procedure allowed a classification of the importance of the studied factors based on their significance in explaining the variance of the measured variables.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Drainage Periods and Amounts
Drainage occurred during three separate periods. In the first winter, drainage began late in the season and lasted until the end of April. After very heavy rain in June 1997, drainage occurred in some of the lysimeters at the end of June and the beginning of July. Lastly, during the second winter, drainage began in December 1997 and lasted until the end of March 1998. Irrespective of the period, drainage varied inversely with N fertilization level because larger dry matter (DM) growth utilized more water. The higher the mineral fertilization rate, the smaller the volume of the drainage recorded (Table 3) and the higher the DM production: for 1996, there was 3, 6, and 8 (±0.2 Mg) DM ha–1 for 0N, 150N, and 300N, respectively, and in 1997, there was 3, 7, and 11 Mg DM ha–1 for 0N, 150N and 300N, respectively. In contrast, as they only slightly influenced the DM production, neither the amount of urinary N nor the date of application had any significant impact on the drainage volume (P > 0.05). Actually, their impact on DM production was much smaller than that of fertilizer application with average values of 6, 5, and 5 (±0.2) Mg ha–1 for spring, summer, and fall urine application, respectively, and 5, 6, and 6 (±0.2) Mg ha–1 for no urine, A2, and A3 urine amounts, respectively, in 1996. This lesser influence of urine compared with fertilizer might result from the small proportion of the lysimeter area able to benefit from the urine N applied on 20% of the lysimeter surface. During the first winter period, the amounts of water drainage remained small (Table 3) but at values typical for this location (long-term average of 100 mm).


View this table:
[in this window]
[in a new window]
  		Table 3. Water drainage amounts according to N fertilizer application and drainage period.

 
Total Nitrogen Leaching Losses during the 1996–1997 Winter
Total amounts of NO–3–N losses during the first winter period remained moderate (between about 2 and 50 kg N ha–1 yr–1), but the variation between treatments was considerable as shown in Table 4. The impact of the three studied factors on these leaching losses is summarized in Table 5. These results allowed the relative importance of each factor to be classified, when studied in combination, according to their statistical significance. The amount of N applied as urine was the main factor influencing N leaching. Nitrogen fertilizer application had a smaller influence, and the date of urine application was not a significant factor in explaining N loss.


View this table:
[in this window]
[in a new window]
  		Table 4. Drainage and leaching data over the two experimental years.

 

View this table:
[in this window]
[in a new window]
  		Table 5. Mean N leaching in the first and second winter as affected by urinary N applied, N fertilizer application, and date of urine deposition.

 
Nitrogen leaching losses result from a combination of the volume of water drained and its nitrate concentration. As differences within N fertilizer treatments in the volume of drainage were small, they did not have a significant impact on total NO–3–N losses in our experiment (P > 0.05). In contrast, the correlation between N loss and NO3––N concentration was very significant (r2 = 0.93). In most treatments, NO–3–N concentration dynamics followed a typical pattern. The first step lasted until the end of January with the concentration of NO–3–N showing either a slight decrease (spring and summer treatments) or remaining stable (fall treatments) (Fig. 1) . After the end of January, the NO–3–N concentration in all urine treatments (except 0N spring) increased significantly reaching peak concentrations within a 5-d period. As drainage progressed, NO–3–N concentrations stabilized or decreased. The first 50 mm of drainage was associated with little N being leached (Fig. 2) . It corresponded to the period when NO–3–N that had accumulated in surface soil layers was moving down through the soil. When this mass of excess NO–3–N reached the bottom of the lysimeters, the drainage contained a large mass of NO–3–N. When all the accumulated NO–3–N was leached, NO–3–N concentration in additional drainage was small and diluted the soil water stock remaining at depth (after mid-March). The 300N fall treatment was an exception to this pattern because drainage stopped earlier (Fig. 1) due to the early (mid-March) growth of the sward, which had much available N from high fertilizer application and N remaining from the fall urine application. Consequently, the peak nitrate concentration did not reach its maximum in the 300N fall A3 treatment (Fig. 1), and we did not observe any end of drainage decrease in nitrate concentration. In all treatments, except those without urine, 0N spring, 0N summer, and 150N spring A2, the mean annual NO–3–N concentration exceeded the potable water threshold of 11 mg NO–3–N L–1 (Table 4).



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 1. Nitrate concentration in drainage water in winter and spring 1996–1997 according to fertilizer level and urine application date. Filled diamonds, no urine; filled squares, total N A2; open squares, urine N A2; filled circles, total N A3; open circles, urine N A3. 0N, nitrogen fertilization of 0 kg ha–1 yr–1; 150N, nitrogen fertilization of 150 kg ha–1 yr–1; 300N, nitrogen fertilization of 300 kg ha–1 yr–1; A2, treatment providing 10.5 g N m–2 as urine; A3, treatment providing 16.5 g N m–2 as urine.

 


View larger version (36K):
[in this window]
[in a new window]
  		Fig. 2. Cumulative N leaching losses. (a) Mean according to fertilizer application: filled triangles, 0N; filled squares, 150N; filled circles, 300N. (b) Mean according to urine N content: filled triangles, no urine; filled squares, urine A2; filled circles, urine A3. 0N, nitrogen fertilization of 0 kg ha–1 yr–1; 150N, nitrogen fertilization of 150 kg ha–1 yr–1; 300N, nitrogen fertilization of 300 kg ha–1 yr–1; A2, treatment providing 10.5 g N m–2 as urine; A3, treatment providing 16.5 g N m–2 as urine. Each point is the mean of three to nine data ± SE.

 
The analysis of variance performed on the mean annual concentrations showed that the amount of N applied as urine was the main factor influencing NO–3–N concentration. Nitrogen fertilizer application had less influence, and date of application did not modify NO–3–N concentration significantly (Table 5). This was confirmed by the step-by-step regression, which indicated that 70% of the variance in NO–3–N leaching losses was explained by the level of urinary N. The addition of N fertilizer level as a factor helped explain a further 14% of the variance (r2 = 0.84), and addition of the date of application to the previous factors explained just 3% of the observed variance (r2 = 0.87).

Figure 2 helps explain the discrepancy between the impact of the different factors on NO–3–N concentration and NO–3–N leaching amounts. We observed that increasing N fertilizer level significantly increased NO–3–N concentration very early in the drainage period, that is during the first 50 mm of drainage (Fig. 2a). However, increased fertilizer application also led to smaller amounts of drainage water. The balance between increased nitrate concentration and decrease in the volume of drainage water caused the impact of the N fertilizer level being less on N leaching losses than on NO–3–N concentration (with P < 0.05 and P < 0.01, respectively) (Table 5). In contrast, the application of urine N did not influence the drained volume during the first winter and spring (Fig. 2b) so that its effect on NO–3–N concentration and NO–3–N leaching was similar (Table 5).

To further assess differences between the application of N as fertilizer and as urine to a sward on N losses, we calculated the increase in N leaching resulting from the addition of one additional kilogram of N in the range of 0 to 30 g m–2 as fertilizer and 0 to 16.5 g m–2 for N as urine, using the data of Table 4. For 1 kg N added to the sward, the increase in N leaching averaged 57 (±25) mg for fertilizer and 170 (±27) mg for urine.

Urinary Nitrogen-15 Leaching Losses during the 1996–1997 Winter
The urinary 15N contribution to total N leaching greatly varied within the urine treatments and ranged from 4 to 94% (Fig. 1). Both 15NO–3–N concentration (Table 6) and 15NO–3–N percentage of total N leaching were significantly influenced by N fertilizer level, date of urine application, and quantity of N applied as urine. However, the interaction between these factors was not significant. Date of application had the most important influence, followed by the quantity of N applied as urine, while fertilization level had the smallest effect (Table 6). Thus, inorganic N leaching was predominantly composed of urinary 15N in the fall and A3 treatments. Contrary to the effect on total N leaching (Table 5) and the concentration of 15NO–3–N (Table 6), the rate of fertilizer application did not influence 15N leaching losses. Date of application was the main driving factor of 15N leaching losses with fall treatments being different from spring and summer treatments, which were similar. Urinary 15N contribution to total N leaching was 36, 64, and 78% for spring, summer, and fall treatments, respectively.


View this table:
[in this window]
[in a new window]
  		Table 6. Urinary 15N leaching losses in the 1996–1997 winter as affected by urinary N applied, N fertilizer application, and date of urine deposition.

 
Very little 15NO–3–N was detected in the 0N spring and the 150N spring A2 treatments. In contrast, 15NO–3–N leaching losses from the urine applied in autumn were about 100 and 500% greater than when applied in summer or spring, respectively. The main reason for the difference was that after spring application, significant grass growth as well as immobilization in the soil utilized most of the 15N applied, whereas little or no plant growth occurred after the fall application. Most urine N was then particularly vulnerable to immediate loss during the winter interval.

Recovery of 15N as percentage of the applied N urine (Table 4) largely varied, but remained moderate reaching 27% as a maximum. Application date was the main factor contributing to the variability in recovery with mean values of 4.33 (±1.7), 12.9 (±2.1), and 21.4% (±1.7) for spring, summer, and fall treatments, respectively.

Nitrogen Leaching during the 1997–1998 Winter
During the second winter, natural rainfall plus additional irrigation led to large mean drainage volumes (Table 3) compared with the previous winter. Drainage amounts were about twice the typical values and represented about double the soil water holding capacity. This enhanced the possibility of previously immobilized N and 15N to mineralize and leach during the second season. Amounts of total N leaching losses remained relatively small (from approximately 2–20 kg ha–1, except for the 300N fall treatments) and mainly depended on N fertilizer level when it reached 300N (Table 5). Mean N leaching losses from the 300N treatments were 2.5 times greater than in 0N and 150N treatments. Date of urine deposition also had a significant effect on the total amount of N leached (Table 5), but there was an interaction with fertilizer application rate. The higher the fertilizer rate, the more the application date influenced N leaching. Nitrogen leaching losses increased by 50% between spring and summer applications, and by a further 50% between summer and fall treatments. The amount of N applied as urine in Year 1 had no influence on losses in the subsequent year (Table 5).

The mass of 15N leached was very small except for the 300N fall treatments. As a percentage of total N leached, urine contribution to leaching depended only on urine application date (4, 11, and 42% for spring, summer, and fall, respectively).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Leaching Characteristics in Winter 1996–1997
Two main observations are usually reported in the literature: (i) the quantity of drainage is a determinant of NO–3–N leaching losses for grass swards (Jordan and Smith, 1985; Jarvis et al., 1989) and (ii) NO–3–N concentration in the drained water significantly decreases at the end of the drainage period (Steele et al., 1984; Scholefield et al., 1993; Silva et al., 1999). We did not observe these effects due mainly to the very small drainage amount (i.e., 100 mm on average) associated with small winter rainfall occurring in our location compared with most values reported previously. Scholefield et al. (1993) in the UK, Heng et al. (1991) in New Zealand, Stout et al. (1998) and Owens et al. (1999) in the USA, and Simon (1995) and Loiseau et al. (2001) in France reported 500, 300, 250, 400, 500, and 600 mm of drainage, respectively. Thus, the small drainage volume in the first winter period resulted in a nonsignificant relation between N leaching losses and the drainage volume in our experiment. Moreover, in many treatments the small drainage did not permit the completion of the typical pattern of NO–3–N concentration evolution during a drainage season (stability, increase, and decrease), the last step being interrupted by the cessation of drainage. However, the peak concentrations observed in our experiment were in the range of previously reported values (155 and 55 mg NO–3–N L–1 for Steele et al. [1984] and Jarvis et al. [1995], respectively). As a consequence of the lack of any decrease in the NO–3–N concentration late in the drainage period, mean annual NO–3–N concentrations were larger in our treatments than values reported by other workers. For example, Mc Duff et al. (1990) and Heng et al. (1991) observed 2.5 and 11 mg NO–3–N L–1, respectively, in similar conditions of grazing and fertilizer application.

Impact of Nitrogen Fertilizer Application, Date of Urine Deposition, and Urinary Nitrogen Amount
Impact of N fertilizer application, date of urine deposition, and urinary N amount whose impact we assessed in our climatic conditions are generally considered to have greatest impact on NO–3–N leaching from grazed grassland. Jarvis et al. (1989), Barraclough et al. (1992), Decau and Salette (1994), and Ledgard et al. (1999) observed that increasing the N fertilization level increased N leaching losses. Cuttle and Bourne (1993), Williams and Haynes (1994), Stout et al. (1997), and Di and Cameron (2002) found that late urine deposition date led to increased N leaching. Williams and Haynes (1994) demonstrated large differences in N leaching beneath a urine patch according to the amount of N applied as urine. In contrast to the results summarized above, with a combination of these three factors and the small drainage usually observed at our location, date of urine deposition had a significant influence on the contribution of urine to total N leaching (which increased with late season deposition) but not on total N leaching losses. Moreover, the apparent contribution of urine remained very low for the fall treatments with the highest observed data of 25%, whereas other studies usually indicated about 60% for late application (Sherwood, 1986; Cuttle and Bourne, 1993; Di and Cameron, 2002). Our results showed that N leaching losses increased with increasing N fertilizer level between 0 to 300 kg N ha–1 yr–1. This increase was particularly significant between 150N and 300N treatments. However, we found that while N fertilization increased NO–3–N concentration, it reduced the drained water volume, and this resulted in a moderate impact on total N leaching. In our conditions, the amount of N deposited as urine was by far the most important factor influencing N leaching. One additional kilogram of N added on the sward through urine led to increase N leaching losses three times more than 1 kg N applied as fertilizer. This much greater impact on leaching of additional N when applied as urine than when applied as fertilizer might have in part resulted from N fertilizer impacts on the drained volume. In addition, it might also be attributable to their deposition characteristics. Fertilizer is usually applied on the sward in several dressings and evenly distributed. On the contrary, urine deposition is applied at one given time throughout the grazing season and only covers a small area of the sward resulting in a concentration both in time and space. Hence, an improvement in the manipulation of supplemental feeding for grazing animals, leading to a reduction in urine N content (Deenen and Lantinga, 1993; Delaby et al., 1997), might be far more effective in improving water quality than the simple reduction of N fertilizer application to grassland.

This research confirmed the recent work underlining the major role of urine in the leaching process under grazed conditions (Hack-ten Broeke and van der Putten, 1997; Di and Cameron, 2002) and confirmed that the concentration of N on a small part of the area enhanced the impact of animal returns (even when the amounts applied were compatible with grass uptake ability).

Urinary Nitrogen-15 Recovery and Contribution to Nitrogen Leaching
The contribution of urinary 15N to total N leaching varied between 4 and 94% among our treatments. The greatest contribution seemed very large considering that only 20% of the lysimeter area received an equivalent of 82.5 g N m–2 when compared with contributions of 80 to 88% reported by Clough et al. (1998) beneath a 100 g N m–2 urine patch. This high contribution of urinary 15N in leachate resulted from the low soil N contribution, and recoveries of the applied 15N in leaching remained in the range observed by other workers (0–26% in our study). Actually, Williams and Haynes (1994) obtained recoveries in leaching of 1 and 11% varying with increasing amount of N applied as urine. Clough et al. (1998) showed that the recovery also varied according to soil type from 13 to 20%.

Nitrogen and Urinary Nitrogen-15 Leaching in the Longer Term
Though in a very different climatic context, our results on N and 15N leaching losses in the second winter period confirmed that N applied as urine has no long-term effect on N leaching as previously reported by Clough et al. (1998), who did not find any 15N leaching in the second year with 848 mm of drainage.

In contrast to expected results, our results for the first year showed that, even with moderate N fertilizer and small amounts of N applied as urine, the mean annual NO–3–N concentration of the drained water exceeded the 11 mg L–1 threshold for drinking water as soon as urine was applied after spring.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
This study clearly demonstrated that, in locations where drainage is poor, the impact of grazing (through urinary N amounts, urine N concentration, and season of urine deposition) on water quality may be much higher than N fertilization management, within the range of 0 to 300 kg N ha–1 yr–1.


   ACKNOWLEDGMENTS
 
is made to the Conseil Régional de Basse Normandie for funding the lysimeters. We are grateful to D. Perrin for his skilled and steadfast assistance with field work. L. Delaby, J.L. Peyraud, and R. Vérité from SRVL, INRA, Rennes deserve special mention for urine collection and valuable advice and comments on the experimental design.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2004 33: 413-418. [Full Text]  




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 Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in 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 Web of Science (2)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Agricola
Right arrow Articles by Decau, M. L.
Right arrow Articles by Jacquet, A.
Related Collections
Right arrow Grazing Management
Right arrow Nitrogen
Right arrow Lysimeter/Rhizosphere Studies
Right arrow Plant and Environment Interactions
Right arrow Water Pollution

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