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


     

Published online 9 August 2006
Published in J Environ Qual 35:1803-1811 (2006)
DOI: 10.2134/jeq2006.0003
© 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 Bergström, L.
Right arrow Articles by Kirchmann, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bergström, L.
Right arrow Articles by Kirchmann, H.
Agricola
Right arrow Articles by Bergström, L.
Right arrow Articles by Kirchmann, H.
Related Collections
Right arrow Water Quality
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Plant and Soil Interactions
Right arrow Animal Waste

TECHNICAL REPORTS

Plant and Environment Interactions

Leaching and Crop Uptake of Nitrogen and Phosphorus from Pig Slurry as Affected by Different Application Rates

Lars Bergström* and Holger Kirchmann

Department of Soil Science, Swedish University of Agricultural Sciences, P.O. Box 7014, SE-75007 Uppsala, Sweden

* Corresponding author (lars.bergstrom{at}mv.slu.se)

Received for publication January 2, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The influence of increasing pig slurry applications on leaching and crop uptake of N and P by cereals was evaluated in a 3-yr study of lysimeters filled with a sandy soil. The slurry was applied at N rates of 50 (S50), 100 (S100), 150 (S150), and 200 (S200) kg ha–1 during 2 of the 3 yr. The P rates applied with slurry were: 40 (S50), 80 (S100), 120 (S150), and 160 (S200) kg ha–1 yr–1. Simultaneously, NH4NO3 and Ca(H2PO4)2 were applied at rates of 100 kg N ha–1 and 50 kg P ha–1, respectively, to additional lysimeters (F100), while others were left unfertilized (F0). During the 3-yr period, the leaching load of total N tended to increase with increasing slurry application to, on average, 139 kg ha–1 at the highest application rate (S200). The corresponding N leaching loads (kg ha–1) in the other treatments were: 75 (F0), 103 (F100), 93 (S50), 120 (S100), and 128 (S150). The loads of slurry-derived N in the S100, S150, and S200 treatments were significantly larger (P < 0.05) than those of fertilizer-derived N. In contrast, P leaching tended to decrease with increasing input of slurry, and it was lower in all treatments that received P at or above 50 kg P ha–1 yr–1 with slurry or fertilizer than in the unfertilized treatment. The crop use efficiency of added N and P was clearly higher when NH4NO3 and Ca(H2PO4)2 were used rather than slurry (60 vs. 35% for N, 38 vs. 6–9% for P), irrespective of slurry application rate. Therefore, from both a production and water quality point of view, inorganic fertilizers seem to have environmental benefits over pig slurry when used on sandy soils.

Abbreviations: a.u., animal units • DM, dry matter • EU, European Union • PVC, polyvinyl chloride


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
DRASTIC CHANGES have occurred within the agricultural sector since the 1950s and farms have become more intensive and specialized, with animal and crop production often being concentrated in certain regions (Bergström et al., 2005). This has led to increased potential for leaching losses of nutrients, of which N and P are of most concern since they can cause severe environmental disturbances in various water bodies (Schindler, 1977; Kirchmann et al., 2002). As a consequence, much legislative and regulatory effort has been devoted to this issue during the last few decades, for example, the U.S. Clean Water Act of 1972 and the more recent European Union (EU) Nitrate Directive, which mandates designation of NO3-vulnerable zones in each EU member state. In parallel, farming strategies to mitigate nonpoint-source pollution have been developed, including use of cover crops and lower fertilization to reduce N losses (Lord and Mitchell, 1998; Bergström and Jokela, 2001), and buffer strips and conservation tillage to reduce P losses (Withers and Jarvis, 1998).

Much of the concern about large diffuse loadings of both N and P on surface waters and ground water is related to animal manure (Patni and Culley, 1989; Smith et al., 1998). On many farms today, the number of animals far exceeds the land base needed to recycle the nutrients contained in the manures in an environmentally satisfactory manner (Sims et al., 2005), and excessive amounts of manure are often applied on agricultural soils, which poses a serious risk for water degradation (Walter et al., 1987). In the EU Nitrate Directive mentioned above, it is therefore stipulated that the maximum amount of manure N that can be applied in a vulnerable zone be limited to 170 kg ha–1 yr–1. In 2001, almost 40% of the total EU land area was designated (or scheduled to be designated) as NO3 vulnerable. Most countries also restrict manure application during certain parts of the year, for example the Netherlands, where manure cannot be applied from 15 September to 1 February. In Sweden, it is not permitted to spread manure on frozen soil (1 January–15 February).

As mentioned above, the rate at which manure is applied on individual fields, which ultimately determines the risk of leaching, is to a large extent determined by the number of animals on the farm. This is commonly regulated by allowing a maximum number of animal units (a.u.) per hectare, which is different in different countries. In Sweden, for example, 1.4 large a.u. (1 large a.u. is equal to one dairy cow or nine slaughter pigs) are allowed, with the intention of limiting the annual applications of manure P to 22 kg ha–1. This is based on the assumption that by setting a maximum P application rate, N will seldom be applied in excess of crop demand, since the N/P ratio of animal manure is often quite low (Eck and Stewart, 1995). However under certain conditions, allowing 1.4 a. u. could cause unacceptable leaching losses of N. For example, in a study in which equal amounts of manure N and inorganic fertilizer N (100 kg ha–1) were applied to barley (Hordeum distichum L.) in spring, leaching loads of N during a 3-yr period were one order of magnitude larger from manure than from fertilizer (Bergström and Kirchmann, 1999). To achieve the same leaching level of manure-derived N as that obtained when 100 kg N ha–1 of mineral fertilizer was used, the number of a. u. had to be lowered to 0.2 ha–1. This stresses the importance of taking extra precautions when determining the application rate of manure to agricultural crops.

The main objective of the study presented in this paper was to quantify leaching and crop uptake of N and P after application of pig slurry at different rates, and to compare the results with those obtained after application of inorganic fertilizers at typical rates in Scandinavian cereal production. An additional objective was to investigate how the repeated application of slurry affected leaching after the soil had been left fallow during the final growing season and sown with a winter cereal crop during the subsequent autumn. The study was performed over a 3-yr period in field lysimeters filled with a sandy soil.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Experimental Design
Eighteen lysimeters filled with a loamy sand/sand (Typic Cryopsamment; Kirchmann, 1985) were used. Standard polyvinyl chloride (PVC) sewage pipes (0.295-m diam., 1.18-m long) were filled with the soil, which was collected from a field site (Pustnäs) located 5 km south of Uppsala, Sweden (59°49' N, 17°39' E) in December 2001. The sandy texture throughout the profile and the low organic matter content, and thus low N mineralization of soil organic matter, were preferred properties for this study in which the dynamics of N release from added manures were the main object of investigation. Loamy sand/sand soils represent about 12% of agricultural soils in Sweden (Eriksson et al., 1999). Selected soil properties are given in Table 1. The soil profile was divided into three layers (0–25, 25–70, and 70–108 cm), which were repacked into the PVC pipes layer by layer in April 2002. Below the soil profile, a 3-cm layer of gravel (2- to 4-mm diam.) was placed on top of a fiberglass lid with five holes (0.01-m diam.) to provide free drainage. To keep the bulk density similar in each layer of the different lysimeters, the same volume of soil from each layer was gently rammed to give layer thicknesses equal to those in the undisturbed soil. The soil columns were then water saturated from the base and thereafter allowed to drain freely, and left to settle for about a month before the experiment was started. Repacking the soil was considered appropriate because it is likely to introduce only small changes in water transport and nutrient behavior in coarse-textured soils such as that used in this study (Bergström, 1990).


View this table:
[in this window]
[in a new window]
  		Table 1. Selected properties of the soil used in lysimeters.

 
A few days before the experiment started, the lysimeters were placed in pipes permanently installed below ground at a lysimeter station in Uppsala, Sweden, 1 km west of the field where the soil was collected. The station and the preparation of lysimeters are described in more detail elsewhere (Bergström, 1992).

All cultivation procedures performed in the lysimeters were performed to simulate field conditions. On 31 May 2002, about 2 cm of soil was removed from the top of the soil profiles, oat (Avena sativa L.) was sown at a rate corresponding to 285 kg ha–1 and then the soil was replaced on top of the seeds. On 16 June 2002, pig slurry was spread at N rates equivalent to 50 (S50), 100 (S100), 150 (S150), and 200 (S200) kg total N ha–1, each rate on three replicate lysimeters. The slurry, which originated from an animal research station in Uppsala, was collected from a slurry tank after mixing in spring 2002 and was stored in a closed plastic container at 1°C before use. The animal diet was based on 80% cereals, 20% pea, oilseed rape, and soybean meal. The slurry had a dry matter (DM) content of 8.2% and a pH (H2O) of 6.8, and contained 0.16% total N (0.10% NH4–N), and 0.13% total P. This resulted in P rates of about 40 (S50), 80 (S100), 120 (S150), and 160 (S200) kg ha–1. The relatively low N content in the slurry, amounting to only about 50% of concentrations normally found in slurry (Steineck et al., 1999), was probably due to large losses during storage. The duration of storage was about 2 mo. The slurry tank was not covered and, in combination with the warm weather during mixing, conditions were favorable for NH3 volatilization. At the same time as the slurry was applied, three lysimeters received NH4NO3 at a rate of 100 kg N ha–1 (F100), and three others received no N fertilizer or slurry (F0). Immediately after the slurry and fertilizer were applied, 3 cm of topsoil was spread on top of the columns to minimize the risk of NH3 volatilization from slurry N. The application of slurry (the same product as used in 2002, which had been stored at –20°C) and N fertilizer was repeated at the same rates on 18 May 2003. On this occasion, about 3 cm of soil was removed on each lysimeter, followed by application of the slurry and fertilizer. Then about 1 cm of the removed topsoil was replaced before barley was sown at a rate corresponding to 285 kg ha–1, after which the rest of the removed soil was added. The NH4NO3 fertilized lysimeters received inorganic P [Ca(H2PO4)2] and K (KCl) at rates of 50 kg ha–1 in both years simultaneously with the N fertilizer application. The oats and barley were harvested on 19 Sept. 2002 and 3 Sept. 2003, respectively. The aboveground plant parts were cut at ground level with scissors and dried in an oven (40°C) to determine the dry mass. In 2004, no slurry or fertilizer was applied and the soil was left fallow during spring/summer. On 6 July 2004, all lysimeters were treated with glyphosate (N-(phosphonomethyl)glycine) at a rate corresponding to 3 L a.i. ha–1, to kill all the weeds that were growing on the soil. Winter wheat (Triticum aestivum L.) was then sown on 24 Sept. 2004 at a rate corresponding to 285 kg ha–1 in a similar way as described above for barley. No data on wheat yields are included in this paper since it was not harvested during the experimental period.

In addition to natural precipitation, which was measured in Uppsala at the Dep. of Earth Sciences, Uppsala University (http://www.geo.uu.se/), all lysimeters received supplementary irrigation via spray bottles 19 times during the experimental period to simulate normal precipitation for the area (554 mm yr–1). On each irrigation event, 7.5 (or once 15) mm of water was added, totaling 150 mm over the period. On each occasion, water was added at rates not exceeding the infiltration capacity of the soil.

Results presented below as yearly amounts refer to the 12-mo periods from 1 June to 31 May of the following year.

Leachate Sampling and Analytical Procedures
Water freely draining through the soil columns was collected in glass sampling bottles placed in the measuring station. The bottles were weighed every 2 wk to determine the leachate volume. If sufficient amounts of leachate were available, samples were then taken for determination of total N and total P concentrations. The loadings of each constituent were calculated by multiplying the leachate volume for each period by the concentrations for that period. Sampling started in 1 June 2002 and continued until 31 May 2005.

To determine the total N concentrations in water samples, inorganic and organic N constituents were oxidized by K2(SO4)2 + NaOH to NO3 (Stevenson, 1982), which was analyzed by flow-injection analysis (Tecator AB, Höganäs, Sweden) according to the colorimetric Cd-reduction method (American Public Health Association, 1985). The total P concentrations were determined on unfiltered samples according to methods issued by the European Committee for Standards (1996).

After harvest, the dried crop samples were separated into straw and grain, which were weighed and milled. Subsamples of each fraction were then analyzed for total N by dry combustion (Leco CNS-2000, Leco Corp., St Joseph, MI; Kirsten and Hesselius, 1983). Total P was analyzed on barley samples by inductive coupled plasma (ICP) (Perkin Elmer, Wellesley, MA) after digestion with concentrated HNO3. Total P contents were not determined for oat grown during 2002, since the differentiated P application rates with slurry were considered to have marginal effect on P concentrations in crop after only 1 yr.

The amounts of N and P in crop and leachate derived from fertilizer and manure were estimated by use of the difference method, that is, a direct comparison with an unfertilized control (Jansson and Persson, 1982). The nutrient use efficiency was then calculated by dividing these amounts of N and P in a harvested crop by the nutrient application rates (Ladha et al., 2005).

Statistical treatment of data on leaching of N and P, crop yield, and uptake of N and P was performed by one-way analysis of variance using the SAS procedure ANOVA (SAS Institute, 1985). Mean value comparisons between the different treatments were made by Duncan's multiple range test and Tukey's studentized range test.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Precipitation and Drainage Conditions
During the 3-yr period (1 June 2002–31 May 2005), cumulative precipitation was 1745 mm, which, in combination with irrigation, resulted in a total water input in the lysimeters of 1895 mm (Table 2). This total water input is about 14% higher than the long-term average precipitation for the Uppsala region. The wettest year was the second year (2003–2004; 678 mm), followed by the first (2002–2003; 621 mm) and the final year (2004–2005; 596 mm). During the peak of the growing season (June and July), the highest precipitation (214 mm) occurred during 2002. Precipitation was also proportionally much higher in May 2003 (92 mm) than in May 2004 (54 mm) and May 2005 (56 mm). In combination with supplemental irrigation, this meant that 52% of the total water input in 2002–2003 occurred during June 2002, July 2002, and May 2003. In the two other years (2003–2004 and 2004–2005) the corresponding figures were 31 and 38%, respectively. The long-term average precipitation in Uppsala during this period is about 30% of total annual precipitation.


View this table:
[in this window]
[in a new window]
  		Table 2. Precipitation and irrigation and mean annual amounts of leachate (±SD; n = 3) during the 3-yr period.

 
The cumulative amounts of leachate each year are shown in Table 2. As expected, the largest amounts of leachate were measured during the wettest second year (2003–2004) in all treatments, which is a reflection of the strong correlation between leachate and precipitation. However, the difference was larger than could be explained by the different water inputs, for example, there was almost twice as much leachate during 2003–2004 compared with the other 2 yr (Table 2). As mentioned above, a larger proportion of the water inputs occurred outside the cropping season in 2003–2004 than in the other years. This contributed to decreased transpiration and thereby increased leachate volumes. A pattern characterized by intensive drainage in unfrozen soil without a crop is typical of northern European climatic conditions (Hansen and Djurhuus, 1996).

The total drainage volume during the 3-yr period was significantly smaller (P < 0.05) in the S200 treatment (414 mm) than in F0 and F100 (about 460 mm) (Table 2). Otherwise there were no significant differences (P > 0.05) in drainage volumes between different fertilizer/slurry treatments (Table 2), which could be expected since such differences are often of the same magnitude as those between replicates of the same treatment (cf. Bergström and Kirchmann, 1999). The smaller drainage volume in the S200 treatment was presumably mainly attributable to higher crop yields and thereby higher transpiration rates by the crop. This could be a likely explanation for the difference between the S200 treatment and the unfertilized control (F0), which had considerably lower crop yields (see below). However, it cannot explain the difference between the S200 and F100 treatments, which had similar crop yields (see below). Another possible explanation for the reduced amounts of drainage could be that the addition of relatively large amounts of organic material with pig slurry in the S200 treatment increased the water-holding capacity of the soil. This is in line with what was suggested by Bergström and Kirchmann (2004) in a study in which N leaching was measured in NH4NO3–fertilized lysimeters and lysimeters receiving green manures, and drainage volumes were significantly reduced (by 20%) in the latter.

On average, leachate volumes were 24% of precipitation plus irrigation over the 3-yr period, which is about the same as found in other similar leaching studies (Bergström and Jokela, 2001).

Crop Yields and Nitrogen and Phosphorus Uptake
Dry matter yields of grain and straw in the different treatments are shown in Table 3. Grain DM yields during the first year (2002) tended to be higher in the slurry treatment with the lowest application rate (S50) and in the NH4NO3 treatment (F100), without being significantly different (P > 0.05) than grain yields in the other treatments. Straw DM yield was highest in the F100, S150, and S200 treatments and tended to increase with increasing slurry application. All fertilized/manured treatments, except the S50 treatment, had significantly higher (P < 0.05) straw yields than the unfertilized treatment. During the second year, all treatments except F0 were fertilized again and slurry N applications equal to or exceeding 100 kg ha–1 increased grain yields to levels similar to those of the F100 treatment. Yield at the lowest slurry application (S50) was significantly lower (P < 0.05) than the other fertilized/manured treatments, except the S100 treatment. Straw DM yields increased with slurry application rates and were significantly different (P < 0.05) from each other.


View this table:
[in this window]
[in a new window]
  		Table 3. Crop dry matter yields (mean values ±SD; n = 3) during 2002 and 2003.

 
Uptake of N in grain was significantly increased (P < 0.05) by the fertilizer and slurry applications in both years (Table 4). Nitrogen uptake at the highest slurry rate (S200) was similar to that of the NH4NO3 treatment (F100).


View this table:
[in this window]
[in a new window]
  		Table 4. Total N removal with crops (mean values ±SD; n = 3) during 2002 and 2003.

 
There was a positive relationship between applied slurry N and total uptake of N in grain and straw. Nevertheless, uptake of N in grain was not as efficient in the slurry treatments as in the F100 treatment during the first year (2002)—every 50 kg of N applied as slurry was equivalent to 30 kg NH4–N. Higher N contents in straw of slurry-treated crops (S150, S200), which amounted to 0.8% of DM compared to 0.5% of DM in the F100 treatment, indicated that slurry-derived N was available at a later stage of crop development and not as efficiently transformed into N in grain as the fertilizer-derived N. The most likely reason is that some inorganic slurry N was initially immobilized and then remineralized during a later stage of plant development (Kirchmann and Lundvall, 1993). However, this N taken up by crops rather late in the season may not be relocated into grain. A similar pattern was observed for the harvest index values (grain weight divided by aboveground biomass weight), also indicating that the N availability during grain formation was limited. In the following year (2003), crop uptake of slurry N was again positively related to the application rate.

Calculating the crop utilization of N applied showed that, on average, 38% of slurry N and 64% of fertilizer N were taken up during the first year after application. During the second year, similar N utilization rates as during the initial year were measured, 37% for slurry N and 61% for fertilizer N. There was only a minor variation in the percentage utilization between the different slurry application rates.

There tended to be positive relationship between applied slurry P and total uptake of P in grain and straw (Table 5). Total P uptake at the highest slurry application rates (120 and 160 P ha–1 yr–1) was similar to that of the F100 treatment, which received Ca(H2PO4)2 (50 kg P ha–1 yr–1). This means that crop uptake of slurry P was not as efficient as uptake of P from the inorganic fertilizer, as N was the limiting factor for crop production at lower slurry application rates. The N/P ratio of the manure was 10:8, which is far from being optimal for grain crops. Phosphorus was simply added in excess through manure in relation to the crop demand.


View this table:
[in this window]
[in a new window]
  		Table 5. Total P removal with barley (mean values ±SD; n = 3) during 2003.

 
Concentrations of P in grain from the high slurry applications (S150, S200) were about 0.4% of DM (data not shown). This concentration is slightly below what is considered a threshold value for fodder (0.45%); higher concentrations in crops than this threshold value often result in nonefficient utilization by animals (Valk et al., 2000). The high P application through slurry to obtain similar yields to those obtained with a fertilizer (N–P–K composition of 100–50–50) indicates that it is difficult to control the P composition in crops when using slurries. Even if the N concentrations in the slurry had been twice as high (less N losses during storage) and thus similar to commonly reported levels (Steineck et al., 1999), there would still be an excess application of P with slurry and it is likely that P concentrations in crops would still be close to or above the given threshold value. Calculating the crop utilization of applied P during 2003 revealed that only 6 to 9% of slurry P was used by the crop depending on the slurry rate, but 38% of fertilizer P.

Leaching of Nitrogen and Phosphorus
Nitrogen concentrations in leachate (Fig. 1 ) tended to be higher in the S200 treatment, reaching on average 66 mg total N L–1. Otherwise, there was little difference in N concentrations between the other fertilized–manured treatments, whereas the unfertilized lysimeters (F0) had the clearly lowest N concentrations in leachate most of the time. In all treatments, the N concentrations were high in autumn each year, when leachate volumes increased, and then tended to decrease during the subsequent spring. A similar leaching pattern has been observed by Kladivko et al. (1991), who attributed the increased NO3–N concentrations in tile-drainage water to large amounts of residual N in soil after harvest.


Figure 1
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 1. Total N concentrations (±SE, n = 3) in leachate on each sampling occasion.

 
Regarding total N leaching loads (Table 6), the largest annual loads occurred during the first year (2002–2003), except in the S50 treatment, even though the drainage volumes were considerably larger during the second year (Table 2). These larger N loads were obviously due to the higher total N concentrations during 2002–2003. One possible explanation could be that the high precipitation during summer (June and July) in 2002 triggered mineralization of organic N fractions in the topsoil, and the inorganic N formed subsequently moved down in the soil profiles beyond reach of the plant roots. When drainage flows started in late autumn and continued during the next spring (2003), the water contained large amounts of N, which leached out. As much as 49 kg total N ha–1 leached on average in the S200 treatment during spring 2003 (January–May). High mineralization rates in response to warm and moist conditions have been shown in several previous studies (Doran and Smith, 1987).


View this table:
[in this window]
[in a new window]
  		Table 6. Average annual leaching loads of total-N (±SD; n = 3) during the 3-yr period.

 
In total over the 3-yr period, the treatments with slurry N applications at or exceeding 100 kg ha–1 yr–1 during the first 2 yr had significantly larger (P < 0.05) N leaching loads than the other treatments, reaching 139 kg ha–1 in the S200 treatment (Table 6). Comparing the slurry treatments, annual leaching of N was positively correlated to the N application rates (R2 = 0.95; P < 0.005) (Fig. 2 ). The smallest N leaching loads were measured in the F0 treatment, which did not receive any fertilizer or slurry at all. Even when the N sources were applied at the same rate (100 kg N ha–1 yr–1), the N load was significantly larger when slurry was applied compared to NH4NO3, which is remarkable since about 40% of N in the slurry was in organic form and not readily leachable. On the other hand, a large proportion of this organic N was most likely mineralized after the cropping season, when drainage occurred, which contributed to the larger leaching loads. A high crop utilization of N in animal manure during the first year after application is a prerequisite for keeping leaching loads at a minimum. Several studies have shown that N leaching loads commonly increase when organic manures are used (e.g., Kemppainen, 1995; Thomsen et al., 1997), due to poor synchrony between the release of N from the manures and the N requirements of crops (Bergström and Kirchmann, 1999).


Figure 2
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 2. Mean annual leaching of N and P (±SE, n = 3) as a function of application rates of N and P with slurry. Slurry was only applied during the first 2 yr at N and P rates shown on the x axis.

 
Over the 3-yr period, leaching of total N derived from fertilizer and slurry, as estimated with the difference method, was: 14% (F100), 18% (S50), 22% (S100), 18% (S150), and 16% (S200) of applied N in the respective treatments. Expressing leaching loads of total N in relation to harvested grain yields, the relative numbers in the different treatments were: 13 (F0), 11 (F100), 11 (S50), 14 (S100), 14 (S150), and 17 (S200). In other words, the slurry treatments with N application rate equal to or exceeding that of the NH4NO3 treatment had 27 to 54% larger N leaching loads per harvested unit than the NH4NO3 treatment.

One of the objectives of this study was to investigate the effect on N leaching of repeated applications of slurry followed by a year with fallow, and to compare this with a system in which the slurry was replaced with inorganic N fertilizer. Fallowing is a common, politically driven practice within EU countries today. When fertilizer was used (F100), the N leaching load was significantly smaller (P < 0.05) during the final year (22.7 kg N ha–1) compared to the second year (38.9 kg N ha–1), whereas in the slurry treatments (except S50) there was no such decrease. In fact, in the S100 treatment the N leaching load tended to be larger during the final year (36.3 vs. 38.2 kg ha–1), although the difference was not significant (P > 0.05). This indicates that there was considerable mineralization of organic N fractions derived from the slurry a year after the slurry had been applied. This agrees well with results presented by Bergström and Kirchmann (1999), which showed that the largest N leaching loads of manure-derived N occurred during the third year after manure application.

Phosphorus concentrations in leachate are shown in Fig. 3 . The highest concentration peak occurred in leachate of the S150 treatment, which reached, on average, 0.22 mg total P L–1, followed by the S100, F0 and S50 treatments, which also had average peak concentrations exceeding 0.1 mg total P L–1. In the slurry treatments, these concentration peaks were associated with low drainage flows (<8 mm), whereas the peak in the unfertilized treatment (F0) (on average 0.15 mg total P L–1) occurred during a period with high drainage flows (>60 mm). Most of these peaks occurred during spring/summer, at the end of the drain-flow period rather than in the beginning when drainage started in autumn, as was the case for N.


Figure 3
View larger version (24K):
[in this window]
[in a new window]
  		Fig. 3. Total P concentrations (±SE, n = 3) in leachate on each sampling occasion.

 
Total P leaching loads over the 3-yr period were overall quite small, not exceeding 250 g ha–1 (Table 7). The F0 treatment had the largest total P load. All other treatments, except the S50 treatment, had significantly smaller (P < 0.05) P leaching loads than the unfertilized treatment (F0). In contrast to N, the largest P loads occurred during the second year (2003–2004) in all treatments, reaching 148 g ha–1 in the unfertilized treatment (F0). Another difference to N leaching was that P leaching tended to decrease with increasing slurry application rates (Fig. 2). This is rather remarkable considering that the S200 treatment received an input of 320 kg P ha–1 during the period, which was clearly in excess in relation to what was needed by the crop (see above). A similar observation was made by Djodjic et al. (2004), who found that in three out of five soils investigated, P leaching loads tended to decrease after P inputs had increased over a long time period. They attributed this result to the potential of the soil to release P, the capacity of the subsoil to adsorb P, and the water transport mechanism. However, none of those explanations is likely to explain the results obtained in the present study, or at least not the last two; the pH of the sandy subsoil was 7.0 (Table 1) and the P sorption capacity was therefore presumably low, and preferential flow was quite unlikely in this sandy soil with hardly any organic matter in the subsoil (Table 1). It is also hard to explain the larger P leaching loads in the F0 treatment by the lower crop uptake of P in this treatment (Table 5), since leaching losses of P occurred in late autumn/spring during intensive drain-flow periods. An explanation for the reduced P leaching loads with increased slurry applications could possibly be a shift in soil P chemistry as suggested by Sharpley et al. (2004). They found that long-term additions of manure increased the pH of the surface soil, shifting P to relative insoluble Ca complexes, which could contribute to reduced P leaching. In another study, Gaston et al. (2003) observed that water-soluble P was 13% of Melich-3P in a sandy loam soil similar to that used in the present study, but the fraction of water-soluble P dropped to 5% after 20 yr of poultry litter application. Both these studies point to a diminished fraction of water-soluble P with long-term manure addition, but we do not know how applicable this is for interpretation of the results obtained in our study, where only two manure applications occurred. However, it may point to one chemical mechanism, namely a general shift from Fe- and Al- to Ca-reaction products. The fact that the response curve for annual P leaching vs. slurry P application rate (Fig. 2) was better fitted to an exponential function (R2 = 0.91) than a linear (R2 = 0.84) indicates that if a component reducing P leaching (e.g., Ca) was added with slurry, the lower slurry application rates (S50 and S100) would be sufficient to decrease leaching of P. Nevertheless, a better understanding of the chemistry regarding forms and solubility of P in manured soils is needed.


View this table:
[in this window]
[in a new window]
  		Table 7. Average annual leaching loads of total-P (±SD; n = 3) during the 3-yr period.

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Results from this study showed that leaching of total N increased with increasing pig slurry applications. At any slurry rate at or above 100 kg N ha–1 yr–1, which was the application rate of NH4NO3 in a comparative treatment, N leaching of slurry-derived N was significantly larger than that of fertilizer-derived N. In contrast, P leaching tended to decrease with increasing input of P with slurry, and it was lower in all treatments that received P inputs at or above 50 kg P ha–1 yr–1 with slurry or fertilizer than in the unfertilized treatment. This suggests that to regulate animal manure application rates based on P inputs is questionable for sandy soils, at least if environmental concerns related to P leaching are the motivation. On the other hand, if the manure application rate is based on N input, it is extremely difficult to avoid luxury consumption of P by crops, leading to an inefficient use when grain is fed to animals.

Out-take of N in grain was largest in the NH4NO3 fertilized and S200 treatments. However, the use efficiency of added N was clearly best if NH4NO3 was used rather than pig slurry, about 60 and 35%, respectively, irrespective of slurry application rate. The use efficiency of added P was also best when an inorganic P fertilizer was used. One can therefore conclude that, from both a production and water quality viewpoint, inorganic fertilizers [NH4NO3 and Ca(H2PO4)2] seem to be superior to pig slurry when used on sandy soils. However, animal manures will certainly be produced in large amounts also in the foreseeable future, and we have to optimize their use so we take advantage of a potentially good nutrient source and avoid losses to the environment. Farms raising animals should have the land base, or access to sufficient land, to recycle manure nutrients in accordance with codes of good agricultural practice. If not, economically viable options for alternative uses for surplus manure should be available. Organic manure nutrients can also be transformed to inorganic form by various processes and thereby be distributed over large areas. The bottom line message is to manage nutrients, whether they are in organic or inorganic form, in a sustainable manner.


   ACKNOWLEDGMENTS
 
This work was funded by a grant from the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (Contract no. 22.0/2001-1117), to which we express our sincere thanks.


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




This article has been cited by other articles:


Home page
 Agron. J.Home page
J. S. Paschold, B. J. Wienhold, D. L. McCallister, and R. B. Ferguson
Crop Nitrogen and Phosphorus Utilization following Application of Slurry from Swine Fed Traditional or Low Phytate Corn Diets
Agron. J., June 16, 2008; 100(4): 997 - 1004.
[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 Bergström, L.
Right arrow Articles by Kirchmann, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bergström, L.
Right arrow Articles by Kirchmann, H.
Agricola
Right arrow Articles by Bergström, L.
Right arrow Articles by Kirchmann, H.
Related Collections
Right arrow Water Quality
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Plant and Soil Interactions
Right arrow Animal Waste

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