Effects of DMPP on the growth and chemical composition of ryegrass (Lolium perenne L.) raised on calcareous soil

This paper reports a 406 day outdoor experiment (performed in pots) to determine the influence of the new nitrification inhibitor 3,4-dimethylpyrazole phosphate (DMPP), when added to pig slurry (PS), on the growth and chemical composition of ryegrass. Pots containing a loamy, calcareous soil were treated with either no PS or 73.7, 147.3 or 221 cm pot, with or without DMPP, prior to seeding with ryegrass. The greatest quantity of above ground dry matter was obtained with the highest dose of PS + DMPP (36.3 g pot) – 7.4% greater than that obtained for the same treatment without DMPP, and 46.1% greater than with the no PS treatment. The plants treated with the high and medium doses of PS + DMPP absorbed 70% of total N during the first quarter of the experimental season (104 days). Without DMPP, N uptakes were 55.7% and 63% for the high and medium treatments respectively. The inhibition of nitrification with DMPP increased agronomic efficiency and reduced N leaching by 17%.

source of N and P for crops, but the amounts applied are sometimes greater than those that might be considered appropriate agronomic doses (20-50 and even up to 200 m 3 ha -1 year -1 ).The Catalonian autonomous government has adopted several measures to prevent the overuse of animal manures in line with European legislation.In 1998 a Code of Best Management Practices was passed (DOGC, 1998), and in 2001 a law was brought into force that obliges farmers to draw up N management plans (DOGC, 2001).The calendar for PS application in Catalonia depends on the crop.For example, in maize f ields, PS is spread between February (two months before the sowing period) and July.The amount of N applied as PS is limited to 210 kg N ha -1 per year (30 to 40 m 3 ha -1 ).
There is strong social pressure to reduce environmental pollution, although farmers also want to diminish the use of nitrates in order to reduce costs.In recent years, technological innovations in the fertiliser industry have sought to produce «eco-fertilisers» that are more compatible with the natural environment.Nitrification inhibitors (NI) are compounds that delay the bacterial oxidation of ammonium ions to nitrite (and subsequently to nitrate) by suppressing the activity of Nitrosomonas spp. in the soil (Prasad and Power, 1995;Trenkel, 1997;Zerulla et al., 2001).NI have been used to increase yields and to reduce nitrate leaching in several crops (Malzer and Randall, 1985;Frye et al., 1989;Malzer et al., 1989;Walters and Malzer, 1990;Serna et al., 1994;Corré and Zwart, 1995;Davies and Williams, 1995;Martin et al., 1997;Trenkel, 1997;Ball-Coelho and Roy, 1999;Pasda et al., 2001).
The addition of NI to different types of slurry has been the subject of numerous studies.Schröder et al. (1993) observed that the addition of dicyandiamide (DCD) to cattle slurry did not sufficiently improve N recovery by silage maize to justify its recommendation, and concluded that the risk of pollution could only be limited by reducing the dose of N to rates below economically optimum levels.DCD has also been added to cattle slurry (Corré and Zwart, 1995), to pig slurry (Tittarelli et al., 1997) and to the urine of dairy cattle (Di et al., 1998).Schmitt et al. (1995) confirmed the results reported by McCormick et al. (1984), indicating that the effect of nitrapyrin on maize yield, when added to different types of manure (dairy and swine), was inconsistent.These yields increased at some sites while no response was seen at others where nitrapyrin was added to manure applied at different rates.Amberger (1990) recommended the addition of a NI to prevent groundwater pollution, especially if slurry is applied in late autumn or winter.
In the USA, nitrapyrin is the most extensively used NI (Walters and Malzer, 1990), but in Europe the most used is probably DCD (Zerulla et al., 2001).
The compound 3,4-dimethylpyrazole phosphate (DMPP) is a new nitrification inhibitor developed by BASF (Germany) in cooperation with a number of universities and research institutes (Conrad, 2000;Zerulla et al., 2001).
The main aim of this work was to assess the agronomic and environmental effects of DMPP when added to PS.The variables studied were the above ground dry matter yield of ryegrass grown in pots, the concentration of nutrients in these plants, the quantity of N volatilised during the growth season, total absorbed nutrients, and the amount of nitrogen leached.A number of nitrogen use efficiency ratios were also calculated.
This work was part of a project to assess the behaviour of DMPP in calcareous soils in nitrate-sensitive areas using field and pot experiments (Carrasco and Villar, 2001).

Material and Methods
The outdoor trial was performed in pots over a period of approximately one year (June 2001 to September 2002).The pots were 25 cm in diameter and 23.5 cm deep, and were filled with 12.25 g of a loamy, calcareous soil taken from the surface layer of an arable area (see Table 1 for soil properties).These pots were assigned to receive one of the following treatments: no PS or either 73.7, 147.3 or 221 cm 3 pot -1 , with or without  , 1991).A volume of 0.02 ml of a 25% DMPP solution was added to the assigned pots.The PS, with and without DMPP, was buried 5 cm under the soil surface 18 days (14 th June) before sowing ryegrass seeds (Lolium perenne L.; 3.26 g pot -1 ; 2 nd July 2001) in the pots.Two more treatments were prepared but without the ryegrass: the low dose PS with and without DMPP.Each treatment was replicated four times according to a completely randomised design.
The PS was collected from a collective slurry storage tank; Table 2 shows its composition.During the experimental period, the total rainfall was 405 mm; 380 mm of water were also supplied by irrigation.
Plant samples were dried for 48 h at 65ºC to determine their dry matter content (yield).Total N was analysed by the Kjeldahl method.Total P and K were determined by the ICP method, but only for the last four harvests.The above ground N, P and K uptakes were calculated by multiplying nutrient concentrations by total plant dry mass.Soil samples were collected on 3 rd September 2002 to determine the residual nutrient content.Leachates were sampled to determine nitrate concentrations at different periods after irrigation.Nitrate concentrations were measured using the Nitrachek ® meter quick test (Merck, Darmstadt, Germany).
The organic matter content of the initial soil sample was determined by the volumetric method (Walkey-Black).Soil pH was measured using a 1:2.5 soil/water weight ratio.P was determined by the Olsen method, using an UV-VIS spectrophotometer.K was extracted using ammonium acetate and measured by the ICP method.Nitrates were extracted using a 1:5 soil/water ratio and colorimetrically analysed using an ICA Autoanalyzer.P and K in PS were determined by the ICP method; total N also was determined by the Kjeldahl method.N-NH 3 volatilisation was measured using open static chambers (with oxalic acid traps); this began on the same day the PS was applied and continued for five days (Teira-Esmatges et al., 2004).
Several nitrogen efficiency indices were used to evaluate plant response to the PS applied and to the nitrification inhibitor (Table 3).The apparent recovery of PS N by the ryegrass (NREC) was calculated as described by Greenwood and Draycott (1989).Physiological efficiency (PE) and agronomic efficiency (AE) were calculated as described by Yadvinder-Singh et al. (2004).
The nitrogen balance was calculated based on the principle of the conservation of mass: N inputs -N outputs = = the change in N content in the pot N inputs were taken as the N levels in the soil at the beginning (N min initial) and at the end of the experiment (N min residual) plus the nitrogen applied with the PS 590 E. Guillaumes and J. M. Villar / Span J Agric Res ( 2004) 2(4), 588-596 (N applied with irrigation was negligible).N outputs were taken to be N uptake, N volatilised after slurry application, and leached N. The N unaccounted for corresponds to mineralised N (input) and N 2 O emissions (output).The data were examined by analysis of variance (ANOVA) and orthogonal contrast.All statistical analyses were undertaken using the Statistical Analysis System (SAS Institute, 1999).

Above ground dry matter yield
The mean dry matter yields were 6.3, 5.6, 3.8, 3.2, 2.7, 4.3, and 3.1 g pot -1 for the seven harvests respectively.A positive response to the application of PS was seen.The accumulated dry matter yield ranged from 24.8 g pot -1 for the no PS treatment to 33.8 g pot -1 for the high PS dose without DMPP treatment (Table 4) -an increase of 36%.The effect of adding the DMPP was also positive.The greatest amount of above ground dry matter was obtained with the highest dose of PS + DMPP (36.3 g pot -1 ), 7.4% greater than the same treatment without DMPP and 46.1% greater than the no PS treatment.The quantity obtained with the medium PS dose + DMPP (33.3 g pot -1 ) was 17.4% higher than the same treatment without DMPP.Finally, the low PS dose with DMPP (28.9 g pot -1 ) produced 17.8% more above ground dry matter than the same treatment without DMPP.Statistical analysis showed the effect PS to be significant (see Table 5).
The inhibition of nitrif ication also resulted in significant increases in yield (Table 5).The differences between the low and the medium and high PS doses were significant, both with and without DMPP.

N uptake and interaction with other nutrients
The application of PS with or without DMPP signif icantly increased N uptake (Table 5).This ranged from 0.78 g pot -1 for the no PS treatment to 1.14 g N pot -1 for the highest PS dose without DMPP (Table 4), an increase of 46%.The increments associated with the high PS dose compared to the low and medium doses were 37% and 21% respectively.The inhibition of nitrification did not influence the total accumulated N uptake.However, when the low dose was compared with the medium and high dose, the differences were even more signif icant when DMPP was present.Moreover, when the velocity of N absorption was analysed, the influence of nitrification inhibition was clearly evident.The high and medium dose treatments of PS + DMPP absorbed 70% of the total N by the third harvest (i.e., during the first 104 days of the experiment), while the treatments without DMPP absorbed 55.7% and 63% respectively (Table 4).
The inhibition of nitrification significantly affected the plant N concentration.This concentration ranged from 3.3 to 3.4% in treatments without DMPP, and from 2.7 to 3.1% in treatments with DMPP.The lower values in the DMPP treatments were due to a dilution effect.In all cases, an increase in yield due to the inhibition of nitrification was associated with a lower plant N concentration.
Phosphorus uptake during the last 302 days of the experiment was significantly greater in the nitrification inhibitor treatments (Tables 4 and 5).No significant effect was observed on potassium uptake.

Nitrogen use efficiencies
The NREC of PS N increased with the amount of N applied, ranging from 13% to 34% in treatments without DMPP (Table 6).These values are relatively low, probably because of initially high soil N levels (Table 1).The low N uptake in the low dose + DMPP treatment (0.77 g pot -1 ) resulted in a negative NREC.The inhibition of nitrification had no effect on NREC (Table 7).NREC increased signif icantly with the medium PS dose + DMPP and the high dose + DMPP compared to the low dose + DMPP.The agronomic efficiency (AE) was similar (10.7 -11.9 g biomass g -1 N uptake) among DMPP treatments.The addition of DMPP significantly affected agronomic efficiency (Table 7).Physiological efficiency (PE) also varied among treatments but the differences were not significant.

Residual nutrients in the soil
The soil N, P and K contents were measured at the end of the experiment (Table 8).Only the no PS treatment had a significantly lower soil N content (Table 9).This  implies that PS had a clear effect on residual N in the soil.The residual amounts of N in the soil at the end of the experiment were 0.11 g N pot -1 on average, 55% less than at the beginning of the experiment (0.25 g N pot -1 ).Similar results were found for residual K. Unlike N and K, the levels of residual P at the end of the experiment did not differ between treatments.

Leached nitrogen
When ryegrass was present, the amount of N leached was significantly lower in the no PS treatment than when PS was added.The quantities leached increased with PS dose (Tables 9 and 10).N leached increased until reaching approximately 15 mm (measured linearly) of the cumulative drainage volume corresponding to the first six leachates.From this point, the nitrate content in the drainage volume diminished abruptly.A dosedependent effect of PS was clear (Fig. 1).
Of the available N, 22.5% was leached in the no PS treatment, 14.5% in the low dose treatment, 16.3% in the medium dose treatment, and 13.4% in the high dose treatment.However, the effect of the nitrif ication inhibitor was only apparent with the high PS dose.The N leached was equivalent to 11.8% of available N for the high dose + DMPP -a reduction of some 15% compared to the high dose without DMPP.The apparent N leached from the PS applied ([N leached fert.crop -N leached for unfert.crop ]*100 / N applied) was 11.9% for the high dose and 9.3% for the high dose + DMPP.
The concentrations of nitrate in the f irst six leachates were high.The highest values ranged from 479 mg NO 3 -L -1 in the no PS treatment to 1,124 mg NO 3 -L -1 in the high PS dose treatment.At the end of the sampling period, nitrate concentrations were below 10 mg NO 3 -L -1 for treatments with ryegrass and above 100 mg NO 3 -L -1 for treatments without the crop.
It is important to highlight the values obtained for the treatments in which there was no ryegrass, especially the low dose PS without DMPP (Table 11).The N leached was 75% of that available (applied + initial mineral soil N content) for the low dose without the ryegrass but only 15% for the low dose with the crop.The N leached was reduced to 63% of the available N for the low dose + DMPP without the crop.This implies a significant reduction of 17% in the N leached when using the nitrification inhibitor (Table 11).

Nitrogen balance
Table 10 shows a simplified N balance.On average, the initial soil N content was 0.24 g N pot -1 (22 ppm NO 3 -N).The available nitrogen (applied + initial mineral soil N content) increased with increasing PS dose.Mean ammonia emissions were 0.003 g N pot -1 (equivalent to less than 1 kg N ha -1 ).The residual N in  the soil was 0.11 g N pot -1 .The levels of unaccounted N depended on the treatments in question.The negative values for unaccounted N imply that some input N was not taken into account, e.g., the mineralisation or denitrification of organic matter.

Discussion
Dry matter yield signif icantly increased with increasing PS dose.This confirms the observations of Adeli et al. (2003) who worked with swine effluent and Bermuda grass.Pasda et al. (2001) performed several field-experiments under different soil and climatic conditions and with different crops, and concluded that DMPP had a positive effect on yields.Possible explanations for the higher yields obtained with NH 4 + -N fertilisers supplemented with NI should be understood in terms of the reduction in N leaching and volatilisation losses, the partial nutrition of plants with NH 4 + , and the improved N supply resulting from fertiliser application.Williamson et al. (1998), who worked with perennial ryegrass, observed an increase in dry matter yield that accounted for the N prevented from leaching due to the use of DCD.
Nitrous oxide emissions produced by denitrification were not measured in this study.Ditter et al. (2001) showed that when DMPP is added to cattle slurry it efficiently reduces N 2 O emissions.Emissions from the slurry pool ranged from 0.93 kg N 2 O ha -1 (without DMPP) to 0.50 kg N 2 O ha -1 (with DMPP), equivalent to 1.45 and 0.78% of the slurry applied respectively.N 2 O emissions can also be reduced by using N fertilisers with DMPP (Linzmeier et al., 2001;Weiske et al., 2001).
The velocity of N absorption was enhanced with DMPP.Williamson et al. (1998) observed that the inhibition of nitrification contributes to the enhancement of plant N uptake.Greater P uptakes due to DMPP (an indirect advantage of its use) might be explained by the fact that DMPP improves the mobilisation and uptake of phosphates from the rhizosphere, probably due to the presence of ammonium ions in the soil solution  (which might reduce soil pH) (Pasda et al., 2001).However, an unknown mechanism may also be at work.Adding DMPP to the medium dose PS significantly increased NREC (from 22 to 31%), but no such effects were associated with other doses.Schröder et al. (1993) observed an increase in NREC following the addition of DCD to cattle slurry.Williamson et al. (1998) found that 32% of the N applied in farm effluent was recovered, compared to 42% of the same effluent when DCD was added.
There were no signif icant differences between treatments in terms of soil residual phosphorous at the end of the experiment because the initial P content was very high (49 ppm, equivalent to 0.543 g P pot -1 ).The quantities of P applied ranged from 0.168 g P pot -1 to 0.508 g P pot -1 .At the end of the experiment, the mean P content was 57 ppm (0.632 g P pot -1 ).
The initial soil K content was low (133 ppm, equivalent to 1.47 g K pot -1 ).The quantities of K applied ranged from 0.327 g K pot -1 to 0.985 g K pot -1 ; at the end of the experiment the average K content was 187 ppm (2.08 g K pot -1 ).
More N was leached in the treatments without the ryegrass than in the treatments with the crop.Beckwith et al. (1998) also reported that the presence of a growing crop reduced N leaching following the application of manure.
As expected, in treatments with the ryegrass, the highest PS dose led to the leaching of more accumulated N.However, if the percentage of leached N is compared to the N applied, the former appears to diminish with increasing PS dose (from 24% to 17%).Nevertheless, inhibiting nitrif ication appeared to have no effect except with the high dose treatment, and even then it was small, as Corré and Zwart (1995) found with DCD.Shepherd (1996), using liquid digested sludge, found that the addition of DCD decreased N leachate losses from the October application.
The effect of PS on the net amount of N mineralised from the soil organic matter was unknown; this represented an important amount of N that remained unaccounted for in the experiment.Unaccounted N (N 2 O emission and N mineralised from organic matter) decreased with increasing PS dose.
Contrary to the «priming effect», the simplif ied apparent N mass balance for the 406 day period shows that the addition of PS (with or without DMPP) seems to reduce the net amount of N mineralised from the soil.
The results of this experiment provide preliminary information on the performance of DMPP in irrigated calcareous soils.Further research is needed to confirm and extend these interesting agronomic and environmental results.

Figure 1 .
Figure 1.Cumulative N-NO 3 -leached in cumulative drainage volume from different treatments.

Table 1 .
Properties of the soil used in the experiment

Table 2 .
Chemical composition of the pig slurry

Table 3 .
Definition of N efficiency indices All variables are expressed in g pot -1 .N applied was considered as total N PS (analysed by the Kjeldahl method).

Table 4 .
Average dry matter yield, average N leaf concentration, total N uptake, and partial N, P and K uptake (Mean ± SD, n = 4) L: low.M: medium.H: high.

Table 6 .
N use efficiency indices: apparent recovery of pig slurry N by ryegrass (NREC), agronomic efficiency (AE), and physiological efficiency (PE).These indices were calculated from accumulated values

Table 7 .
Statistical analysis using contrasts

Table 8 .
Residual soil N, P and K at 7 th October 2002 ND: no data.

Table 10 .
Nitrogen mass balance for a 406 day period expressed as g N pot -1

Table 11 .
Accumulated N leached expressed as g N pot -1 Values followed by the same letter do not differ significantly (P < 0.1) (Duncan's multiple range test).