JEQ Journal of Natural Resources and Life Sciences Education
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 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 (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Agricola
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Related Collections
Right arrow Sustainable Agriculture
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Ecological Risk Assessment
Right arrow Nutrient Management
Right arrow Animal Waste
Published in J. Environ. Qual. 33:1662-1674 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Ecological Risk Assessment

Changes in Soil Nitrogen and Phosphorus under Different Broiler Production Systems

Sylvia Kratz*, Jutta Rogasik and Ewald Schnug

Institute of Plant Nutrition and Soil Science, Federal Agricultural Research Center (FAL), Bundesallee 50, D-38116 Braunschweig, Germany

* Corresponding author (sylvia.kratz{at}fal.de).

Received for publication December 16, 2003.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
In a field study, soils of four conventional free-range and organic broiler runs were analyzed for N and P concentrations in the years 2000 and 2001. Zones of different use intensity by broilers were identified on the free runs and mean zonal nutrient contents were compared with each other. Intensity of use by birds and spatial distribution of soil nutrient concentrations were found to be related to each other. Fecal N input by broilers resulted in accumulation of soil mineral nitrogen (Nmin) contents down to a 90-cm sampling depth. In highly frequented "hot spots," plant requirement as defined by the German "N-Basis-Sollwert" (110 kg/ha Nmin) for grassland was exceeded in all four cases. This implies an increased environmental risk of ammonia volatilization and nitrate leaching. Fecal P input by broilers resulted in accumulation of plant-available and thus mobile soil P (phosphorus extracted with calcium-acetate-lactate [PCAL] and phosphorus extracted with water [Pw]) in the most intensely used zones. In these areas, soil P contents exceeded 90 mg/kg PCAL (upper limit of soil test P defined in Germany for optimum plant yield) by as much as 217 mg/kg, which indicates an enhanced risk of P loss from the soil via runoff or leaching. The conclusion might be drawn that, with regard to nutrient loss from free-run soils, intensive indoor production in a closed system may be more environmentally neutral than conventional free-range or organic production. However, to put this into perspective, the scope of the environmental risk connected with spatially limited point accumulation of nutrients should be considered. Furthermore, an environmental evaluation must also account for the fate and environmental effects of the broiler litter produced inside the broiler house.

Abbreviations: F1 and F2, conventional free-range broiler farms • Nmin, mineral nitrogen • O1 and O2, organic broiler farms • PCAL, phosphorus extracted by the calcium-acetate-lactate method • Pw, water-extractable phosphorus • TC, total carbon • TN, total nitrogen


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
CONVENTIONAL FREE-RANGE and organic broiler production are welcomed as a promising strategy to produce broiler meat in terms of animal welfare issues. At the same time, offering a free run to broilers is sometimes questioned from an ecological point of view (Menke and Paffrath, 1996). Feed concentrates for broilers are rich in nutrients such as nitrogen and phosphorus. However, considerable amounts of the nutrients fed are not utilized by the birds but are excreted with their feces (Kratz et al., 2004). A part of these fecal nutrients is deposited outside the broiler house by birds roaming on the free run. As has been demonstrated for pigs (Brandt et al., 1995), laying hens (Menke and Paffrath, 1996), and turkeys (Haneklaus et al., 2000), this will lead to enrichment of N and P in the topsoil. Model-based calculations for phosphorus (Kratz, 2002) showed that up to 60% of the total amount excreted by broilers during a fattening period is deposited in the free run. Other than in conventional broiler farming, there is hardly any chance to control amount and distribution of feces on the free run. Since space is not used evenly by birds, nutrient input will be concentrated in areas preferred by them. Preferential usage of protected areas close to the barn is a particular problem with broilers, because they do not live long enough (only 8–12 wk) to get adapted to external disturbances and changing weather conditions in the free run (Menzi et al., 1997). Menzi et al. (1997) have calculated the fecal nutrient input on the most preferred area of a broiler free run to be about 11 times higher than the mean input on the whole free run. Fecal N and P inputs on free runs may reach or already exceed nutrient uptake by an intact grass cover even at a modest frequency of use (i.e., 25% of feces produced during grazing period excreted outside on the free run) by broilers (Kratz, 2002).

Overloading soils with N and P bears the risk of ammonia volatilization from the soil into the atmosphere as well as N and P leakage into surface or ground water, with subsequent acidification and eutrophication of soil and water. While many studies exist on environmental effects of broiler litter application to farmland (Liebhardt et al., 1979; Schilke-Gartley and Sims, 1993; Sharpley et al., 1993; Kingery et al., 1994; Sims and Wolf, 1994; Lucero et al., 1995; Gordillo and Cabrera, 1997; Marshall et al., 1998; Vervoort et al., 1998; Bergström and Kirchmann, 1999; Sharpley, 1999; Williams et al., 1999; Sauer et al., 2000), there is only little information available on changes of soil chemical properties related to fecal nutrient input by broilers on free-run soils. In this paper, results are reported from a field study performed on two conventional free-range and two organic broiler farms. The aim of the study was to:


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
In the years 2000 and 2001, soil samples were collected from one organic (O1) and two conventional free-range (F1, F2) broiler farms with fixed location of barn and free run, and one organic farm with mobile wooden sheds and flexible fences rotating on several plots (O2). Organic farms in this study were defined as farms producing in accordance with EU Regulation 1804/1999 or with private standards set by German organic farmers' associations. Conventional free-range farms were defined by EU Regulation 1538/91 together with 2891/93.

All birds had access to the free run from Growing Day 28 until slaughter. Main differences between farms were the size of the free run, number of broilers, space offered per bird, and duration of fattening period (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Main characteristics of conventional free-range (F) and organic (O) broiler farms studied.

 
Sampling Strategy
A variable soil sampling strategy was followed taking into account the specific conditions prevailing on the free run, in particular the distribution of birds in space, which was influenced by local microclimate and vegetation cover. A first survey was performed in a coarse grid (grid dimensions varying between 5 and 50 m, depending on total size of free run) to determine mean nutrient status of the soils in spring 2000 (sampling depth = 0–90 cm, in 30-cm increments). During the grazing period of 2000, at each farm samples were taken at the beginning and end of one fattening period. In areas used more intensively by birds, additional sampling points were established (sampling depths = 0–10 and 0–30 cm). Due to financial and time limitations, this extended grid could not be fully sampled at all sampling dates. Therefore, in autumn 2000 and spring 2001, only selected points in highly and minimally frequented areas were sampled to a 90-cm depth (and 0–10 cm in spring 2001). A complete survey with 0- to 10- and 0- to 30-cm sampling depths was done again in autumn 2001, with selected points sampled down to a 90-cm soil depth as well (Table 2).


View this table:
[in this window]
[in a new window]
  		Table 2. Details of soil sampling in free-range (F) and organic (O) broiler farms.

 
As it was impossible to exactly quantify the amount of feces deposited in the free run, nutrient contents in the 0- to 10-cm soil depth were used as an indicator of fecal nutrient input. This sampling depth was selected for two reasons. First, the idea was to describe changes in the upper soil layer induced by fecal nutrient input. With a shallower sampling depth, the risk of strong contamination by concentrated feces would have been increased. Second, a <10-cm sampling depth would have been restricted to the main rooting zone of the grassland soil. Both feces and roots would have impaired the representativity of the obtained soil samples.

To study the influence of broiler movements on soil nutrient contents, areas of different frequency of use (Zone 1 = highest frequency of use, Zone 2 = medium frequency, Zone 3 = low frequency, and Zone 4 = very low frequency or no use at all) were defined based on visible characteristics, such as:

Soil nutrient contents were interpreted by comparing zonal means.

Chemical Analyses
From an environmental perspective, plant-available and thus easily soluble nutrient contents are the most important fractions, since they are most likely to be lost from the soil via runoff or leaching (Schoumans et al., 1997; Schoumans and Groenendijk, 2000). Therefore, soils were characterized for plant-available nutrient contents using the following methods. For mineral nitrogen (Nmin), 150 g field fresh soil were shaken with 300 mL 0.0125 M CaCl2 solution for 1 h and then filtered using two Macherey-Nagel (Düren, Germany) 614 1/4 filters (Verband Deutscher Landwirtschaftlicher Untersuchungs-und Forschungsanstalten, 1987). For nitrate, microbiological detection after Kücke and Przemeck (1982) was performed; for ammonium, colorimetric determination was done based on the Berthelot reaction with indophenol after Houba et al. (1986), using a Nanocolor test set. For both determinations, a PerkinElmer (Wellesley, MA) 550SE UV/VIS spectrophotometer was used. Phosphorus extraction with calcium-acetate-lactate (PCAL) was done according to Schüller (1969). Five grams of air-dried soil, passed through a 2-mm sieve, were shaken for 2 h with 100 mL calcium-acetate-lactate (CAL) solution. Phosphorus extraction with water (Pw) was done according to Van der Paauw et al. (1971). Air-dried soil (1.5 g) was passed through a 2-mm sieve, moistened with 2 mL deionized water, and left for 22 h at 20°C (room temperature). After adding 70 mL of deionized water, samples were shaken for 1 h and than filtered with four Schleicher & Schuell (Dassel, Germany) 593 1/2 filters. Colorimetric determination of both PCAL and Pw was performed on the extracts with molybdenum blue (Murphy and Riley, 1962) using a PerkinElmer 550SE UV/VIS spectrophotometer.

To quantify the influence of site-specific soil parameters on the fate of N and P in the soil, pH, total nitrogen (TN), total carbon (TC), and soil moisture were determined using the following methods. Soil pH was determined with a pH electrode in a 0.01 M CaCl2 suspension, where 10 g air-dried soil, passed through a 2-mm sieve, was suspended in 25 mL CaCl2 solution. For TN determination, a Kjeldahl extraction was performed according to Verband Deutscher Landwirtschaftlicher Untersuchungs-und Forschungsanstalten standards (Hoffmann, 1991) using a BÜCHI Labortechnik (Flawil, Switzerland) Model 430 digestor. Extracts were distilled in a BÜCHI 322 distillation unit with 8 M NaOH. The NH3–N was determined by titration with 0.025 M H2SO4 in 0.32 M boric acid, with a pH electrode serving as indicator. Total C was determined by combustion with O2, with Fe and Cu chips as catalysts, using a LECO (St. Joseph, MI) Model 752-100 EC-12 carbon analyzer. Soil moisture was determined in the drying oven, with 50 g field-fresh soil dried at 105°C for 48 h (until mass consistency was reached) (Deutsches Institut für Normung, 2000). Full data for these parameters are given by Kratz (2002).

Statistical Analyses
Mean values were calculated for the total area of the free run as well as separately for each zone of use. For the total area, weighted means were calculated to take into consideration the different relative share each zone had in the total area. For spatial representation of soil nutrient data (Fig. 3), a cluster center analysis was performed with SPSS 8.0 (Brosius, 1998). Differences between zonal means were statistically tested performing a one-way ANOVA (SPSS 8.0) at a significance level of P < 0.05. To determine which means were significantly different from each other, a Bonferroni post hoc test (Bärlocher, 1999) was performed. To describe the influence of site-specific soil parameters on N and P concentrations, multiple regression equations were calculated. Standardized coefficients of regression (ß) were given to compare the relative weight of each parameter in the equation (Brosius, 1998).



View larger version (71K):
[in this window]
[in a new window]
  		Fig. 3. Development of spatial variability of phosphorus extracted with calcium-acetate-lactate (PCAL) concentration in mg/kg (0–10 cm) in the free run of Farm F1. Zone 1/2 is shaded in dark gray, Zone 3/4 in light gray; data were grouped into classes by cluster center analysis and classes are indicated by dot size.

 

   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Mineral Nitrogen
Broiler feces contain about 3.5 to 10% of total N in the form of mineral nitrogen (Kratz, 2002). However, about 40 to 70% of total N is in the form of uric acid, which is rapidly mineralized (Frenken, 1989). Thus, spatial variability of fecal nutrient input should be mirrored by Nmin contents of soil in the short term. Since climatic parameters such as temperature, precipitation, and wind have a strong effect on mineralization, volatilization, and leaching of nitrogen and thus influence variability of soil Nmin contents in time, chronological sequences are not expected to clearly mirror fecal nutrient input by birds. However, differences between zones of use at a particular time may be interpreted with regard to this factor. A spatial differentiation of Nmin contents following frequency of use by broilers was found in all four free runs investigated in this study (Fig. 1). Extremely high Nmin contents were found in the most intensively used areas (Zone 1) of the two organic farms. On Farm O1, Zone 1 corresponded to a trench-like, wind-protected area in front of the barn exits, which most of the broilers preferred to the adjacent open pasture. On Farm O2, this zone corresponded to the area below the wooden shed. The shed was without a floor, so the manure was accumulating on the bare soil during the whole fattening period. On Farms F1 and F2, differentiation between the delimited areas was most clearly visible in the summer months of 2000 (T1 and T2), while differences in Nmin contents in spring (T4, Farm F2) and autumn 2001 (T5, Farms F1 and F2) were not significant. The change from T3 to T4 may be due to flock management on both farms: during the cold winter months, broilers only had access to the free run when temperatures were greater than 0°C. Thus, only a small amount of feces was deposited outside. Volatilization and leaching may have evened out differences existing the year before. On the free run of F2, a new grass seed was established in Zone 1 in late summer 2001, where grass cover had previously been destroyed by broilers. At the time of sampling in October 2001 (T5), the new flock had not yet entered the free run, so the grass cover had been able to grow well. This may have leveled off differences at T5. For Farm F1, no obvious explanation exists. However, it can be assumed that short-term leaching effects due to high precipitation may have overlapped differences in fecal nutrient input at T5.



View larger version (40K):
[in this window]
[in a new window]
  		Fig. 1. Differentiation of mineral nitrogen (Nmin) in different zones of use (as indicated by Zones 1–4) in the free-run soils of conventional free-range (F) and organic (O) broiler farms, 0- to 90-cm sampling depth (O2: 0–30 cm). Significant differences in the 0- to 30-cm soil depth at a particular date are marked with letters. Different scales are due to different orders of magnitude.

 
Figure 1 shows mean Nmin contents for the zones defined according to frequency of use. Mean values exceeded plant uptake in many cases; however, extreme values ("hot spots") were found in the most frequented areas, where fecal nutrient input was considered to be the highest. In the 0- to 30-cm soil depth, maximum Nmin contents ranged between 200 and 298 kg/ha Nmin in F1 and 187 and 728 kg/ha Nmin in F2. In O1 and O2, maximum values were 223 to 847 and 2648 to 2712 kg/ha Nmin, respectively. It should be noted here that by giving these numbers as contents in kg/ha, which is required to evaluate them from an agronomical point of view, an upgrading of point data (concentration) into spatial data (contents) was performed. However, such high amounts of nitrogen are limited to intensively used hot spots only and are not expected to be found over an extended area in the free run.

Depending on the site, between 13 and 47% of Nmin in the 0- to 30-cm soil depth was NH4–N, while 53 to 87% was NO3–N. As can be seen from the correlation matrix (Table 3), different sampling depths were highly correlated with each other. Due to the high leachability of NO3–N, Nmin accumulations in the topsoil continued downward to a 90-cm soil depth (Fig. 1). On average, the NO3–N concentration in the 0- to 30-cm depth was found to be about 76% of the NO3–N accumulated in the 0- to 10-cm soil depth (Fig. 2). Regression equations calculated for 0- to 30-, 30- to 60-, and 60- to 90-cm soil depths revealed that in the 30- to 60-cm depth, soil nitrate contents were about 69% of the topsoil contents (Table 4). The lower coefficient of regression for Farm F1 may be explained by the generally lower Nmin contents of the F1 soil, and thus a lower nitrate leaching potential in that free run. In the 60- to 90-cm soil depth, nitrate contents were about 53% of the 30- to 60-cm-layer contents (Table 4).


View this table:
[in this window]
[in a new window]
  		Table 3. Correlation matrix for NO3–N concentrations and mineral nitrogen (Nmin) contents in different soil depths (N10 for 0–10 cm, N30 for 0–30 cm, N60 for 30–60 cm, and N90 for 60–90 cm) of free runs (data from all four farms).

 


View larger version (20K):
[in this window]
[in a new window]
  		Fig. 2. Nitrate N concentrations in the topsoil (0–30 cm) of four free-run soils (F = conventional free-range farm, O = organic farm) as a function of NO3–N concentrations in the 0- to 10-cm sampling depth.

 

View this table:
[in this window]
[in a new window]
  		Table 4. Nitrate N concentrations in 30- to 60- (N60) and 60- to 90-cm (N90) sampling depths as a function of nitrate concentrations in the topsoil (0–30 cm, N30) and 30- to 60-cm depth, respectively, for conventional free-range (F) and organic (O) farms.{dagger}

 
It can be summarized that fecal nutrient input by broilers resulted in an accumulation of soil Nmin contents down to a 90-cm sampling depth. The Nmin contents in the 0- to 90-cm depth were two to nine times greater in the most frequented zone of F1 and F2 compared with the least frequented zone. In Zone 1 of Farm O1, Nmin contents were even as much as 28 times higher than in Zone 4 (in O2, a layer of hardened clay enriched with large amounts of rocks did not allow for sampling deeper than 30 cm). These results concur with earlier studies, which demonstrated the effect of fecal N input of laying hens (Menke and Paffrath, 1996) and of poultry litter application on grassland (Sharpley et al., 1993; Kingery et al., 1994) on soil nitrate concentrations down to a 100-cm sampling depth.

For an ecological evaluation of fecal N inputs by broilers, Nmin contents in the 0- to 30-cm depth can be related to plant N requirements, which are defined in Germany as "N-Basis-Sollwert" (Schilling, 2000). Kerschberger and Franke (2001) give an amount of 80 to 110 kg/ha N as N-Basis-Sollwert for grassland (0–30 cm) at the beginning of the vegetation period. This number describes the amount of plant-available N (Nmin) needed by the crop to produce an optimum yield. At the same time, it implies that soil Nmin exceeding this content cannot be utilized by plants and thus is susceptible to volatilization and leaching. Of course, at times before or after the vegetation period, when plants do not take up N, or at places where there are no plants to take up N, any N input will be susceptible to these processes. Thus, the risk here was strongly increased by the fact that high Nmin contents in this case were related to intensely used areas with little or no plant cover, which could have taken up the fecal N input.

Soil Nmin contents alone do not allow a quantitative risk assessment. However, it is known from studies on farmland fertilized with broiler litter that up to 60% of the N input with litter may be released into the air as ammonia within a few days (Cabrera et al., 1993). While ammonia emissions are an environmental risk for the atmosphere, nitrate endangers ground water quality. There is a correlation between N input by organic fertilization and nitrate concentration in the ground water (Liebhardt et al., 1979; Bergström and Kirchmann, 1999).

Phosphorus
Phosphorus is subject to runoff and leaching loss to a much smaller degree than nitrate and, in contrast to ammonia, it is not volatized into the air. Therefore, it can be expected that broilers' movements will not only be traceable by soil P contents in space, but also in time. With regard to environmental implications, this hypothesis was tested by analyzing plant-available (PCAL and Pw) soil P concentration, using the concentrations in the 0- to 10-cm soil depth as an indicator of fecal P input by broilers.

Calcium-Acetate-Lactate-Extractable Soil Phosphorus
Figure 3 demonstrates the spatial differentiation of soil PCAL concentrations in the 0- to 10-cm soil depth over a period of 16 mo as a result of frequency of use by broilers. Where the free run was used with medium to high intensity by broilers (Zone 1/2), "hot spots" with increasing soil P concentration developed over time. On the other hand, little or no change in soil PCAL was detected in the less frequented areas (Zone 3/4). A differentiation between zones of use was observed at all four free runs sampled (Fig. 4). However, fecal nutrient accumulation in time was overlapped by a number of natural and artificial processes (Fig. 4). The most obvious feature in all cases was the seasonal variation in soil PCAL, which can be traced back to natural processes such as plant uptake during the vegetation period and mineralization in early spring. On Farm F2, only the most intensely used Zone 1 was distinguishable from the other zones of use. However, this was already the case at the first sampling date (T1), after only one fattening period. No accumulation of P in time was detectable. This may be explained by several factors. In late summer 2001, a new grass seed was established where grass cover had been destroyed by broilers. The growing plants may have taken up part of the fecal P input for buildup of biomass. The reduction in soil PCAL from spring 2001 (T4) to autumn 2001 (T5), however, indicated that, due to the very high soil PCAL concentration already prevailing since the beginning, P sorption capacity of the soil was probably saturated and thus some of the fecal P input was transferred down to deeper soil layers. However, this alone would not account for a difference of about 100 mg/kg between sampling dates. An additional loss of PCAL may have occurred due to P immobilization (Schoumans and Groenendijk, 2000). Plant uptake and P loss probably also leveled off differences between Zones 2, 3, and 4. In addition, interpretation of the data from Farm F2 is hampered due to the strong variability of soil PCAL concentration inside the zones themselves, which is a result of the previous use of this plot as pasture and, furthermore, a spatially irregular fertilization with broiler litter in former years.



View larger version (44K):
[in this window]
[in a new window]
  		Fig. 4. Development of phosphorus extracted with calcium-acetate-lactate (PCAL) in free-run soils (F = conventional free-range farm, O = organic farm) for the total area and separated by frequency of use (as indicated by Zones 1 to 4), 0- to 10-cm sampling depth. Significant differences between zones of use at a particular date are marked with letters.

 
The free run of Farm O1 presented a similar picture: the most intensely used Zone 1 had a significantly higher soil PCAL concentration. However, at T5, soil PCAL in this zone dropped below its initial concentration. In this case, the reason can be found in the artificial trench-like structure bordering the barn exits. Due to erosion processes at the edge of that trench, fecal deposits on the floor were partly covered by sand from time to time. This complicates the interpretation of soil PCAL concentrations in that zone. The weak differentiation of soil PCAL in Zones 2 to 4 can be explained by the intact grass cover prevailing here, which was obviously able to utilize any excess fecal P input in Zones 2 and 3.

Despite very high total amounts, differences in PCAL between zones of use on Farm O2 were often not statistically significant. As in F2, this was an effect of the strong variability inside the different zones. This farm practiced a plot rotation with mobile huts. Therefore, between T1 and T5, the plot was used for broilers only once, and there was a 9-mo ley period after T1 that left the plot unused. Apparently, further rise of soil PCAL on this plot from T1 to T5 could be diminished by the intact grass cover, which had established during the ley period even where the mobile hut had been standing the year before.

Water-Extractable Soil Phosphorus
While spatial variability of soil P was discussed using the PCAL fraction, the water-extractable fraction is important for a risk assessment regarding the loss of P via runoff and leaching. High soil PCAL concentrations were correlated with high water-extractable (Pw) concentrations (r = 0.706 for 0–10 cm and 0.76 for 0–30 cm, both significant at the 0.01 probability level). At the end of the observation period, in autumn 2001, extremely high Pw concentrations were found in the zones of medium to high frequency of use at the free runs of Farms F1, F2, and O2 (Table 5). This was not the case for O1. Apart from the influence of a sand cover coming down from the trench walls in Zone 1, P leaching may play a role here. As the topsoil, and thus any organic substance and metallo–organic complexes, was taken off when creating the trench, the surface soil layer in this area consisted of pure sand, bearing an increased leaching potential. In F2, the differentiation between zones was traceable even down to a 90-cm sampling depth (Table 6). Regression equations were calculated to explain Pw concentrations in deeper soil layers as a function of Pw concentrations in the topsoil (Table 7). While equations for F1 and O2 only explained less than 35% of the variability or were not significant, Pw concentrations in 30- to 60-cm and 60- to 90-cm soil depths of Farm F2 were explained by Pw in the topsoil with a coefficient of determination of R2 = 0.87 and 0.77, respectively (P < 0.001). However, it must be noted that the main reason for extremely high Pw concentrations in the topsoil in this case (Farm F2) was not the short-term use as free run for broilers, but the long-term fertilization with broiler litter, as concentrations were already high at the first time of sampling even at areas least frequented by birds (Fig. 4, Zone 4). Downward movement of P in soils fertilized with broiler litter is a phenomenon that has already been observed in earlier studies (Sharpley et al., 1993; Kingery et al., 1994; Vervoort et al., 1998).


View this table:
[in this window]
[in a new window]
  		Table 5. Water-extractable soil phosphorus concentrations (Pw) and soil test levels (Landwirtschaftskammer Hannover, 1993) in free-run soils of conventional free-range (F) and organic (O) farms at the end of a vegetation period (autumn 2001) at 0- to 10- and 0- to 30-cm sampling depths.

 

View this table:
[in this window]
[in a new window]
  		Table 6. Water-extractable soil phosphorus (Pw) concentrations in the free run of Farm F2 (mean values for different zones of use) at different sampling dates and soil depths.

 

View this table:
[in this window]
[in a new window]
  		Table 7. Water-extractable soil phosphorus (Pw) concentrations in deeper soil layers (Pw60 for 30–60 cm, Pw90 for 60–90 cm) of free-run soils of conventional free-range (F) and organic (O) broiler farms as a function of Pw concentrations in the topsoil (0–30 cm, Pw30).{dagger}

 
Ecological Relevance of Phosphorus Inputs
As was shown in the above sections, fecal nutrient inputs by broilers resulted in plant-available P concentrations (0- to 10-cm sampling depth) in the most intensely used zones being about 1.4 to 2.8 times higher than at the least frequently used zones of the four free runs studied. The results demonstrated that, as a consequence of fecal P input by broilers, very high accumulations of soil P developed in spots preferred by birds in less than two years (Table 8). Basically, this was found true for conventional free-range farms as well as for organic farms. The reason for this was that in both systems, use of the free run by broilers concentrated on fairly small areas close to the barn or protected by shrubs or shading devices, while, according to farmers' and authors' observations, more than 85% of the space offered was only used rarely or not at all. Plot rotation with ley periods or reseeding of damaged grass cover could only dampen, but not completely eliminate, P accumulation in hot spots.


View this table:
[in this window]
[in a new window]
  		Table 8. Accumulation of phosphorus extracted with calcium-acetate-lactate (PCAL) in the highly frequented zones of the free runs of conventional free-range (F) and organic (O) farms over a period of 14 to 16 mo (summer 2000 to autumn 2001).

 
As can be seen from Fig. 4, PCAL concentrations in the highly frequented zones were generally greater than 100 mg/kg PCAL. A threshold value of 90 mg/kg has been defined in Germany as the upper limit of soil test P required for optimum plant yield (Kerschberger et al., 1997). Above this concentration, there is no direct relation between further P input and plant uptake, which implies that any extra P added will be subject to an enhanced risk of loss via runoff or leaching (Kerschberger et al., 1997). (Officially, soil test P must be determined in the 0- to 30-cm soil depth. The 90 mg/kg threshold is only used here to help assess the order of magnitude of the P concentrations found on the free runs).

The main site characteristics governing the degree of runoff from a field are soil type, vegetation cover, aspect, plot slope, and plot length (Beegle et al., 1998; Eckert et al., 1999; Sauer et al., 2000). Where the terrain is flat and soils are well drained, leaching may become the primary route of P loss (Leinweber et al., 1997; Schoumans and Groenendijk, 2000). The actual risk of P loss with leaching water is determined by hydrological parameters such as precipitation and position of ground water table as well as P sorption capacity and degree of P saturation in topsoil and deeper soil layers (Lookman et al., 1995, 1996; Leinweber et al., 1997; Leinweber, 1999). While the scope of this study did not allow us to determine these factors to quantify the actual risk of P loss, looking at the water-extractable P fraction allows a rough risk assessment. This fraction is directly related to the dissolved inorganic P in the soil (Williams, 1966; Moore et al., 1998; Schoumans and Groenendijk, 2000). There are strong correlations between degree of P sorption saturation of a soil, water-extractable P in the topsoil, and concentration of dissolved inorganic P in surface runoff (Pote et al., 1996). Therefore, Pw is viewed as a suitable indicator of the amount of dissolved inorganic P susceptible to runoff and leaching (Schoumans et al., 1997; Moore et al., 1998; Sims, 1998).

At the end of the study period, in autumn 2001, Pw concentrations in the zones frequented with medium to high intensity by broilers were greater than 30 mg/kg Pw in the 0- to 30-cm sampling depth in F2 and O2, which is the lower limit of the extremely high level of Pw (Level E) as defined in Germany (Landwirtschaftskammer Hannover, 1993). As explained before, at Level E, there is much more soluble P present in the soil than required and taken up by plants. Consequently, there is an enhanced risk of P loss. Schoumans and Groenendijk (2000) modeled the concentration of dissolved inorganic P in the soil solution as a function of soil Pw concentrations. They found that Pw concentrations of >50 mg/kg Pw, which were found in the topsoil of F2 and O2, corresponded to an amount of 1 to 2.5 mg/L ortho P in the soil solution. This is 10 to 25 times as high as the limit of 0.1 mg/L ortho P defined in the Netherlands as maximum acceptable environmental P loss (Schoumans and Groenendijk, 2000). If P concentrations in the topsoil solution stay above this limit due to high soil Pw concentrations in that layer, the soil solution of deeper soil layers could also exceed it in the long term.

Influence of Other Soil Parameters on Nitrogen and Phosphorus Concentrations
With regard to vertical movement of N and P in the soil, different regression coefficients for the four free runs studied indicate that, in addition to N and P input by broiler feces, site-specific soil parameters may influence this process, too. To quantify this influence, multiple regressions were calculated for the topsoil (0- to 10- and 0- to 30-cm sampling depths), applying a stepwise exclusion of parameters with P > 0.1. In the case of nitrate, NO3–N and NH4–N in the 0- to 10-cm depth were taken as indicators for fecal N input, and TN, TC, pH, and soil moisture in 0- to 30-cm sampling depths were included as site-specific parameters in the calculations. As TN and TC were highly intercorrelated (r = 0.89, significant at the 0.001 probability level), two separate equations were tested including either TN or TC. As is shown in Tables 9 and 10, NO3–N in the 0- to 10-cm depth was the dominant factor influencing nitrate concentration in the 0- to 30-cm depth, while the effect of TN, TC, pH, and soil moisture was small or negligible.


View this table:
[in this window]
[in a new window]
  		Table 9. Influence of NH4–N and NO3–N in the 0- to 10-cm sampling depth and total nitrogen (TN), pH, and soil moisture (SM) in the 0- to 30-cm depth on soil nitrate concentration in the 0- to 30-cm depth at the free runs of conventional (F) and organic (O) broiler farms.

 

View this table:
[in this window]
[in a new window]
  		Table 10. Influence of NH4–N and NO3–N in the 0- to 10-cm sampling depth and total carbon (TC), pH, and soil moisture (SM) in the 0- to 30-cm depth on soil nitrate concentration in the 0- to 30-cm depth at the free runs of conventional (F) and organic (O) broiler farms.

 
Two regression equations were calculated for PCAL and Pw, respectively. The PCAL or Pw in the 0- to 10-cm depth were taken as indicators for fecal P input, and pH and TC in the 0- to 30-cm depth were included in the equations as site-specific parameters. Tables 11 and 12 clearly demonstrate that PCAL and Pw in the 0- to 10-cm depth were the main factors influencing PCAL and Pw concentration in the 0- to 30-cm depth. Again, the effect of pH and TC was small or negligible in most cases.


View this table:
[in this window]
[in a new window]
  		Table 11. Influence of phosphorus extracted with calcium-acetate-lactate (PCAL) in the 0- to 10-cm sampling depth, total carbon (TC), and pH in the 0- to 30-cm depth on PCAL concentration in the 0- to 30-cm depth at the free runs of conventional (F) and organic (O) broiler farms.

 

View this table:
[in this window]
[in a new window]
  		Table 12. Influence of water-extractable phosphorus (Pw) in the 0- to 10-cm sampling depth, total carbon (TC), and pH in the 0- to 30-cm depth on Pw concentration in the 0- to 30-cm depth at the free runs of conventional (F) and organic (O) broiler farms.

 
Statistical Interpretation
As can be seen from Fig. 1 and 4, differences between zonal means were not always statistically significant despite being large in total amounts. This was the result of high spatial variability of nutrient contents inside the different zones of use. As the number of samples taken from each zone was limited by financial and practical considerations, this variability could not be accounted for. A much larger number of samples would have been necessary to allow for an adequate statistical analysis of the subplots. However, the data give an idea of the trends to be expected from plots used as broiler pasture over an extended period of time.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
In this study, intensity of use by birds and spatial distribution of soil nutrient concentrations were found to be related to each other. Fecal N input by broilers resulted in differentiation of soil Nmin contents down to a 90-cm sampling depth. In highly frequented "hot spots," plant requirement was exceeded in all four cases. Fecal P input by broilers resulted in accumulation of plant-available and thus mobile soil P (PCAL and Pw) in the most intensely used zones. In these areas, soil P contents by far exceeded optimum nutrient supply. The results presented here indicate that offering a free run to broilers is not without ecological risks. If a permanent run with fixed location is used by a large number of birds, "hot spots" with extremely high N and P concentrations may develop due to the irregular spatial distribution of birds and bird litter. High soil mineral N content may give rise to N volatilization and leaching even in a short time. In the long term, P saturation effects of the soil may lead to P discharge into surface and ground waters by runoff and leaching, too. From this point of view, one might come to the conclusion that intensive indoor production in a closed system, which allows for controlled disposal of animal wastes, may be more environmentally neutral than conventional free-range or organic production. However, two aspects must be mentioned here to attenuate this impression. First, this situation is a consequence of the separation of broiler production from farmland and from the on-farm nutrient cycle. Looking at the environmental effects of broiler feces, the fate of the broiler litter produced inside the broiler house must also be considered. According to nutrient balances calculated by Kratz (2002), a broiler farm with 40000 birds produces about 1400 to 1900 kg N and 350 to 470 kg P per fattening period in the form of broiler litter. With seven fattening periods, this amounts to 9800 to 13300 kg N and 2450 to 3290 kg P per year. It is a well-known phenomenon in regions with intensive animal production that farmers are often not able to use these nutrients on the farm but have to give their manure to a waste management facility for proper disposal. Crop farmers receiving the manure from the management facility are in many cases hundreds of kilometers away from the place of its origin (Windhorst, 1996, 1999). Regarding the amount of fossil energy required for transportation, the sustainability of that type of practice is at least questionable (Fleischer, 1998). Thus, understanding agriculture as the cultivation of agricultural lands, an integration of animal and plant production on the farm is desirable. The cultivation of farmland (including animal pasture) as part of a broiler-producing farming enterprise is also necessary with regard to the preservation of traditional agricultural landscapes. This is considered an important aspect of sustainable agriculture (Linckh et al., 1997) and is thus a future task of high public interest.

Second, the scope of the environmental risk connected with a free run should be considered. If average PCAL concentration is calculated over the total area of the pasture, changes in time are negligible apart from seasonal oscillations (Fig. 4). Large accumulations of fecal nutrients are mainly limited to so-called hot spots, that is, to small areas preferentially used by broilers.

With regard to these hot spots it is crucial, however, that farmers and researchers continue to develop alternative solutions to make conventional free-range or organic production more environmentally neutral. The ideal strategy would be the fattening of small groups of only a couple of hundred birds in mobile sheds on rotating plots. However, as the examples of this field study indicate, plot rotation must be practiced with quite extensive ley periods and damaged grass cover should be reseeded regularly. Since this may not be economically feasible for many farms in the short term, more efforts should be made to motivate broilers to evenly distribute in space and thus reduce the spatial concentration of fecal nutrient inputs in particular spots. Important steps toward this goal, such as creating protective structures and shading by planting trees and shrubs, seeding high-growing crops (corn, sunflowers), or putting up mobile sheds and nets, have been presented by Swiss scientists (Hirt et al., 2001; Zeltner and Hirt, 2001).


   ACKNOWLEDGMENTS
 
The field study presented in this paper is part of the joint project, "Interdisciplinary Comparison of Broiler Production Systems of Different Intensity with Special Reference to Ecology, Product Quality, Animal Welfare and Economics." The authors would like to express their gratitude to Prof. Dr. F. Ellendorff for initiating the project, Dr. J. Berk for coordination, and M. Wolf-Reuter and A. Redantz for excellent team work. Special thanks are given to Heinz Lohmann Stiftung, Visbek, for financial support.


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


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2004 33: 1589-1599. [Full Text]  



This article has been cited by other articles:


Home page
 Agron. J.Home page
M. P. Russelle, M. H. Entz, and A. J. Franzluebbers
Reconsidering Integrated Crop-Livestock Systems in North America
Agron. J., February 6, 2007; 99(2): 325 - 334.
[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 Related articles in JEQ
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 (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Agricola
Right arrow Articles by Kratz, S.
Right arrow Articles by Schnug, E.
Related Collections
Right arrow Sustainable Agriculture
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Ecological Risk Assessment
Right arrow Nutrient Management
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