Abstract: The de-inking process produces a waste by-product, called de-inking paper sludge (DS), that contains paper fibers, clay particles and inks and high carbon (C) concentrations combined with low nitrogen (N) and phosphorus (P) concentrations. The use of high rates of DS to increase the soil organic matter thus requires provision of high rates of N and P for adequate plant growth. Using dinitrogen (N2)-fixing forage legumes is an alternative to N fertilization under such circumstances. In a greenhouse study, DS rates of 0, 50 or 100 Mg ha-1 and five rates of P (40, 80, 120, 160, or 200 kg P2O5 ha-1) were applied on two soil types, a clay loam (Pintendre) and a silty clay loam (St-Augustin). Nitrogen uptake and symbiotic N2 fixation (SNF) were estimated in alfalfa (Medicago sativa L.), sweetclover (Melilotus officinalis L.) and red clover (Trifolium pratense L.); Bromegrass (Bromus inermis L.) and alfalfa ineffective for N2 fixation were used as the reference (non-N2 fixing) crops. Atmospheric N2 fixation was estimated by natural abundance of 15N (δ15N). Under controlled conditions, high rates of DS substantially reduced δ15N values, particularly with high rates of P. In addition, N uptake of legumes generally increased with increased P concentrations and it peaked with 120 or 160 kg P2O5 ha-1. Correlated with the trends observed with δ15N values and it peaked with 120 or 160 kg P2O5 ha-1. Present results showed that under high rates of application of DS and adequate P supply, forage legumes fixed more atmospheric N2. δ15N can be a good indicator of SNF under the above-mentioned conditions.
INTRODUCTION
Novel, economically feasible and environmentally sound solutions are required to optimize land use, because of the many demands placed on the world by an ever increasing food requirement (Barrett-Lennard et al., 1986). Applying industrial wastes, such as pulp and paper sludge on agricultural lands, has been found to increase the chemical and physical properties of soil (Sauerbeck, 1987; Araújo et al., 2007). The de-inking process produces a waste by-product, De-inking Sludge (DS), that contains mainly paper fibers, clay particles and inks, which may be very beneficial as potential organic soil amendments and conditioners (Barclay, 1991; NCASI, 1991; Quebec Ministry of the Environment, 1984). The application of DS on fields has to be adjusted according to the exact composition of the sludge, the types of soil present and the type of crop being grown. The chemical composition of the sludge can vary considerably due to the origin of the recycled paper and different processes used by the paper industry (NCASI, 1984). Generally, the concentration of Ca in DS is high and macronutrient concentrations are relatively low, ranging from 0.14% to 4.1% N and 0.001-2.54% P, on dry-mass basis (NCASI, 1984; Beauchamp et al., 1998).
Therefore, a potential limitation for the use of fresh DS as soil amendment is the possibility of N and P deficiencies for adequate plant nutrition and growth, according to its relatively high C:N and C:P ratios which cause N and P immobilization by microorganisms (Tisdale et al., 1993; Fierro et al., 1999; García et al., 2003). The role and requirements of phosphorus with regards to maintaining healthy plants are well documented. Marschner (1995) observed a greater increase in nodule dry weight relative to shoot or root dry weight by increasing phosphorus availability. In dinitrogen-fixing plants, inadequate phosphorus supplies have been linked with nitrogen deficiencies (Dadson and Acquaah, 1984; Fabre and Planchon, 2000). In a greenhouse study (Fierro et al., 1997), all legume species, grown on soil-DS mixtures, showed an increase in shoot DM as P concentration increased.
Under greenhouse conditions, medic (Medicago truncatula) plants assimilated 80% more soil N than wheat at a high rate of P (150 kg P2O5 ha-1) and soil residues incorporation (Elabbadi et al., 1996). In their study, phosphorus treatments increased significantly the amount of nitrogen derived from atmosphere.
Although several studies have examined paper sludge mixtures in crop production (Chong et al., 1987; Chong, 1993; Chong and Cline, 1993; Fierro et al., 1997), the information is scarce on the effect of P rates on growth and symbiotic dinitrogen fixation of forage legume crops under high rates of application of DS without supplemental N, which was the first objective of the present study (Hansen and Vinther, 2001). On the other hand, the lack of fast and precise analytical techniques limited the use of the natural abundance of 15N and it has so far only been used in a limited number of experiments (Bergersen et al., 1990; Ledgard and Peoples, 1988; Peoples et al., 1991; Sanford et al., 1994; Shearer and Kohl, 1986; Hogh-Jensen and Schjoerring, 1994; Wanek and Arndt, 2002). Therefore, the second aim of the present study was to determine the contribution of dinitrogen fixation in forage legumes by the natural abundance of 15N under high rates of application of DS and greenhouse conditions.
MATERIALS AND METHODS
Experimental sites: This study was performed at greenhouse of Laval University, Québec (QC), Canada. The altitude, longitude and latitude, at which this greenhouse experiment was done, are 73 m, 71° 48´ 56´´ W and 46° 38´ 09´´ N, respectively.
Soil fertilizer and sowing application: In November 1995, soils were collected from the surface layer (0-25 cm) of two experimental fields from a freely drained, Silty Clay Loam Soil (SCLS), Typic Dystrochrept relatively low in organic matter (ca. 19.4 g C kg-1) and from an imperfectly drained Clay-loam Soil (CLS), Typic Humaquept with an average of 27.0 g C kg-1. The de-inking sludge was obtained from Les Produits Forestiers Daishowa Ltée' located in Quebec City. Samples of soil and sludge were oven-dried (at 60°C for 72 h) to determine the total and available mineral nutrients (Table 1). Prior to mixing, the samples were oven-dried (at 105°C for 48 h) to determine their moisture concentration.
The soils from the two sites were sieved (5 mm) prior to mixing with paper
sludge. Then, de-inking sludge was passed through a Wiley Mill without screen
prior to mixing with soils to homogenize particle size. Shredded sludge was
mixed thoroughly with the soil at the rates of 0, 260 or 520 cm3
per pot [equivalent to 0, 50 or 100 Mg DS (DM) ha-1] (Table
2).
| Table 1: | Some physical and chemical characteristics of de-inking sludge and soils used in this study. Values are means of three determinations±standard error of means |
| †, Electrical conductivity | |
| Table 2: | The different levels of the sludge-soil mixture in pot experiment |
| †DS, de-inking sludge | |
Different rates of phosphorus (P) were supplied by applying KH2PO4 before sowing (60% of total) and one month after planting (40% of total) at rates equivalent to 40, 80, 120, 160 or 200 kg P2O5 ha-1. Potassium chloride (KCl) was applied at varying rates to maintain potassium (K) concentration similar in all treatments. Top irrigation with tap water was effected to allow only minor leaching that was contained in the saucer; therefore, no mineral nutrients were lost by leaching.
Prior to the preparation of substrate mixtures, a composite sample of each of the basic substrate components was made to determine the available P, K, Ca and Mg (Mehlich III, CPVQ; 1993); totals of P, K, Ca, Mg, aluminum (Al), Fe, Zn and Cu (acid digestion, Olsen and Sommers, 1982) and totals of C and N by dry combustion (CNS-1000 analyzer, LECO Co., St. Joseph, MI). Soil pH and electrical conductivity were determined in a 3:1 (v:v) water:substrate slurry (Sparks et al., 1996) data shown in Table 1.
The forage legumes Medicago sativa (L.) cv. Saranac (alfalfa), Trifolium pratense (L.) cv. Florex (red clover), Melilotus officinalis (L.) cv. Norgold (sweetclover) and the non-fixing reference crops Bromus inermis (L.) cv. Saratoga (bromegrass) and Medicago sativa (L.) cv. Saranac (ineffective alfalfa) (Barnes et al., 1990) were selected for this study.
Legumes were inoculated before sowing with either Sinorhizobium meliloti for sweetclover and alfalfa (De Lajudie et al., 1994) and Rhizobium leguminosarum biovar trifolii for red clover (Jordan, 1984).
Then they were seeded at an equivalent rate of 600 seeds m-2 [11 seeds pot-1 (12 kg ha-1)] in 1400 cm3 plastic pots (14.5 cm diameter by 14.5 cm depth) with saucer, filled with: one of two soil types (silty clay loam or clay loam), mixed with one of three rates of DS equivalent to 0 (DS0), 50 (DS50), or 100 (DS100) Mg dry matter ha-1 and receiving one of 5 P rates equivalent to 40 (P40), 80 (P80), 120 (P120), 160 (P160), or 200 (P200) kg P2O5 ha-1. The substrate mixtures as well as some physical and chemical characteristics of DS and basic components of substrate mixtures are shown in Table 1 and 2, respectively.
Sampling and preparation of plant material: Natural photoperiod at the beginning and at the end of this experiment were 10.1 and 13.7 h, respectively. A photoperiod of 16 h was provided with high pressure sodium lamps (PPFD = 200 μmol m-2 sec-1; PL Light System Canada Inc., Beamsville, Ontario, Canada) and day/night temperatures of 26/18°C were maintained in the greenhouse during 70 days.
At the end of experiment (i.e., 70 days after planting), all above ground plant material was clipped in each pot and dried in a forced-air oven at 60°C until constant weight. The plant samples were finely ground with a Retsch Centrifugal Mill Model ZM-1 (Brinkmann Instruments Canada Ltd., Rexdale, Ontario, Canada) fitted with a 0.5 mm ring sieve. About 10 g of representative subsamples were further ground with a ball mill (Retsch Mixer Mill Model MM-2; Brinkmann Instruments Canada Ltd., Rexdale, Ontario, Canada) to obtain the powder sample and analyzed for percent 15N. The preparation of 15N samples was done with the precautions necessary when measurements of small differences in the abundance of 15N are to follow (Peoples et al., 1991; Eriksen and Hogh-Jensen, 1998).
Determination of dinitrogen fixation: Values of natural abundance of 15N (δ15N; Eriksen and Hogh-Jensen, 1998) of above-ground plant material was also estimated to determine their trends on N2 fixation dependency in the presence of DS.
Measurement of 15N-natural abundance: There are small variations in the natural abundance of 15N and 15N-natural abundance values for the majority of biological systems are within the range 0.3630-0.3700 atom percent 15N (Letolle, 1980).
The relevant conversion formula is:
The following formula is used to calculate the proportion of plant nitrogen derived from atmosphere (Ndfa) by using natural variations in 15N:
where nfs refers to a non-N2 fixing plant selected to match closely the studied legume in terms of uptake of soil sources of N, fs refers to N2-fixing system (nodulating legume) and factor B refers to the δ15N value of the effectively nodulated legume grown in media totally lacking combined N (Unkovich et al., 1994). In this experiment, bromegrass and ineffective alfalfa were used as the reference crops.
Due to the experimental conditions, i.e., short duration of experiment under greenhouse conditions, low quantity of soil (and thus, soil N) in pots, coherent estimates of Ndfa % could not be obtained by 15N-natural abundance method. Thus δ15N values of plant shoots were used as indicator of relative dependency on N2 fixation for the various treatment combinations.
Determine B values of red clover, alfalfa and sweetclover:
Effects of host plant on δ15N of mentioned species solely dependent on fixed N2 (B value) were assessed in minus-N silica (80% silica and 20% perlite) culture conducted in a growth chamber, each plant species comprising four 7 inch pots of 8 plants (inoculated with the appropriate rhizobia for each of two growth-periods (45 or 60 days). The nutrient solution contained macronutrients and trace elements (Chalifour and Nelson, 1987).
At 45 or 60 days after planting, shoots were harvested, bulked on a per pot basis, dried, ground and analyzed for δ15N as described previously.
Temperature regimes in the growth chamber involved nighttime (8 h) minima of 16°C and daytime (16 h) maxima of 25°C.
The average B values obtained from plant materials clipped 45 or 60 days after
planting, were 0.33% for sweetclover, -0.24
Statistical analyses: The experiment was a three-factor factorial in a Randomized Complete Block Design (RCBD) with four replicates. The treatments consisted of three rates of de-inking sludge (for each soil type) and five rates of P (for forage species: red clover, alfalfa, sweetclover, bromegrass and IN alfalfa). For determining the percentages of N derived from atmosphere, only the three specified N2-fixing species were considered. The General Linear Models Procedure (GLM) of the SAS statistical package (Release 6.12, Statistical Analysis System Institute Inc., 1996) was used to test the significance of the associations between each dependent variable and the treatments. Interpretation of statistical analyses was done on interactions, when these were significant. Scatter plots of the residuals from the respective statistical models as well as Bartlett's test (Steel and Torrie, 1980) were used to test homogeneity of the experimental error variances and to determine if data transformations were required. Due to heterogeneity of pooled error variances between the soils of the two sites for nearly all of dependent variables (Gomez and Gomez, 1984), statistical analyses were effected and presented separately for each soil type. Differences among treatments were determined by simple, first-order and second-Order class and trend contrasts (Little and Hills, 1978). F-values were considered significant at the 10% level as described by Steel and Torrie (1980) for the small experiments.
RESULTS
Effect of DS and P rates on N uptake of forage species: On both soil types, forage species showed differential responses in N uptake with DS application (DS x Species, Table 3 and 4) which were mainly due to the differences between N2-fixing and non-N2 fixing species and among N2-fixing crops. Moreover, on the CLS, the effects of P rates on N uptake were not similar among forage crops (Species x P, Table 3 and 4), which were mainly due to the different effects between N2-fixing and non-N2 fixing crops and also between alfalfa and sweetclover.
For both soil types, N uptake of non-N2 fixing plants decreased
strongly with DS application (DS x Species, Table 3 and 4).
On the CLS, N uptake of bromegrass and IN alfalfa were reduced by DS. Bromegrass
N uptake was affected similarly by DS50 and DS100, whereas
N uptake of IN alfalfa continued to decrease at DS100 (DSQ
x Bromus vs IN Alfalfa, Table 3 and 4).
There was a slight increase in N uptake of bromegrass with P addition, at DS0
and DS50 particularly, while no such increase was observed for IN
alfalfa (Bromus vs IN Alfalfa x PL, Table 3 and
4). On the other hand, on the SCLS, N uptake of bromegrass
and IN alfalfa were reduced similarly by DS (Table 3 and 4).
On the CLS, N uptake of N2-fixing plants were not so strongly reduced
compared to non-N2 fixing plants (Table 3 and 4).
| Table 3: | Analyses of variance for nitrogen uptake of forage species as affected by different rates of de-inking sludge and phosphorus on clay loam (Pintendre) and silty clay loam (St-Augustin) soils |
| † df degree of freedom, ‡ ns, not significant, § DSL, Linear effect of de-inking sludge, ¶ DSQ, Quadratic effect of de-inking sludge, ††IN alfalfa, ineffective alfalfa, ‡IPL, Linear effect of phosphorus, §'PQ, Quadratic effect of phosphorus, ¶¶CV, Coefficient of variation | |
De-inking sludge affected N uptake of alfalfa and red clover similarly, while sweetclover was the least affected (DSQ x Alfalfa vs Red clover: non significant, DSQ x Alfalfa vs Sweetclover, Table 3 and 4).
On the CLS, while P had very slight or no effect on N uptake of non-N2
fixing species, P increased N uptake of N2-fixing species and alleviated
the inhibitory effect of DS on N uptake Table 3 and 4).
The enhancement of N uptake by P was higher in red clover than in alfalfa (Alfalfa
vs Red clover x PL, Tables 3 and 4). Because N uptake of sweetclover
was slightly or not reduced by DS, the response to P addition was smaller than
for alfalfa or red clover (Alfalfa vs Sweetclover x PL, Table
3 and 4). On the other hand, on the SCLS, N uptake did
not vary with P addition.
| Table 4: | Nitrogen uptake Medicago sativa L. (alfalfa), Trifolium pratense L. (red clover), Melilotus officinalis L. (sweetclover), Bromus inermis L. (bromegrass) and Medicago sativa (IN) (ineffective alfalfa) as affected by different rates of de-inking sludge and phosphorus on clay loam (Pintendre) and silty clay loam (St-Augustin) soils. Values are means of four determinations±standard error of means |
| †DS, De-inking sludge, ‡Not enough sample for nitrogen analysis | |
Symbiotic N2 fixation: Natural abundance of 15N (δ15N values) of above-ground biomass (shoots) was also estimated, to determine trends in the N2 fixation dependency of forage legumes in the presence of DS.
Effect of DS and P rates on natural abundance of 15N of forage species: On the SCLS, high rates of P generally decreased δ15N of N2-fixing plants, but mostly at DS0 and not at DS50 or DS100; the decreases generally peaked at P120 or P160 (Table 5 and 6). The decreases in δ15N values upon P addition indicate an enhancement of N2 fixation by P. Furthermore, N2 fixation was more strongly enhanced by the application of DS than by P addition (Table 5 and 6). In alfalfa, δ15N values generally decreased strongly at P160 and with increasing DS rates compared with the other P rates, while in red clover, the reducing effect peaked at P160 and DS0 and peaked at P120 and DS100. In sweetclover, δ15N values decreased and peaked at P120 and in the presence of DS50 and DS100.
For both soil types, DS affected the δ15N of forage species differently, mainly according to the different effects between N2-fixing species and reference crops (DS x Species, Table 5 and 6). The presence of DS led to stronger decreases in δ15N values in reference crops than in N2-fixing species for both soils [DSL x Reference vs Fixing crop (both soils) and DSQ x Reference vs Fixing crop (CLS), Table 5 and 6]. With reference crops, the presence of DS caused stronger decreases in δ15N values of IN alfalfa than those of bromegrass [DSL x Bromus vs IN alfalfa (both soils) and DSQ x Bromus vs IN alfalfa (SCLS), Table 5].
On the SCLS, the responses of δ15N to P addition were dependent
on DS rate (DS x P, Table 5 and 6). This
could be described by differential responses between N2-fixing crops
and reference crops. In general, decreasing effect of P on δ15N
values of N2-fixing species peaked at P160 (with DS0
and DS50) or P120 (with DS100), whereas such
responses were not observed in reference crops. Also, on the SCLS, the responses
of δ15N varied among species and P addition (Species x P, Table
5 and 6), which were mostly due to different responses
of bromegrass and IN alfalfa to P addition on the SCLS (Bromus vs IN Alfalfa
x PL, Table 5) with stronger decreases in δ15N
values of IN alfalfa than those in bromegrass.
| Table 5: | Analyses of variance for 15N-natural abundance (δ15N) of above-ground biomass of forage species as affected by different rates of de-inking sludge and phosphorus on clay loam (Pintendre) and silty clay loam (St-Augustin) soils |
| †df, Degree of freedom, δ15N, Natural abundance of 15N, ‡DSL, Linear effect of de-inking sludge, ¶IN alfalfa, ineffective alfalfa, §'DSQ, Quadratic effect of de-inking sludge, ‡IPQ, Quadratic effect of phosphorus. §'CV, Coefficient of variation, ††PL, Linear effect of phosphorus; ns, not significant | |
| Table 6: | Totals of 15N-natural abundance of above-ground biomass of Medicago sativa L. (alfalfa), Melilotus fficinalis. (sweetclover), Trifolium pratense L. (red clover), Medicago sativa (IN) (ineffective alfalfa) and Bromus inermis L. bromegrass) as affected by different rates of de-inking sludge and phosphorus on clay loam (Pintendre) and silty clay loam (St-Augustin) soils. Values are means of four determinations±standard error of means |
DISCUSSION
Effect of DS and P rates on N uptake of forage crops
Non-N2 fixing species: On the CLS, N uptake of bromegrass
and IN alfalfa were reduced by DS. The N uptake of bromegrass increased with
P treatments and P application alleviated the inhibitory effect of DS on N uptake.
On both soil types, decreasing effect of DS on N uptake by reference crops could
be mainly due to the high C:N ratio of DS, low N availability and therefore,
N deficiency with high rates of application of DS in non-N2 fixing
species. Similar trends were found by Feagley et al. (1994), who reported
that N uptake by bermudagrass (Cyanodon dactylon L.) decreased with increases
in papermill sludge rate.
N2-fixing species: For both soils, N uptake of N2-fixing species did not decrease strongly compared with reference crops. On the other hand, on the CLS, P application increased N uptake of these species and again, P alleviated the inhibitory effect of DS on N uptake by N2-fixing species and it peaked at P120 or P160. These results compare well with those of a greenhouse experiment with medic (Medicago truncatula) and wheat (Triticum turgidum) using different rates of P on the soil in which wheat residues (45.5 g plant material pot-1) were incorporated (Elabbadi et al., 1996; García et al., 2003). In their experiment, P treatments increased uptake of soil N for both species.
In contrast, the results of a greenhouse study (Fierro et al., 1997) on the effect of different rates of de-inking sludge and P on the growth of the legumes Galega orientalis, Medicago lupulina and Melilotus officinalis indicated no significant differences in shoot N concentrations, including unamended soil.
Symbiotic N2 fixation
Effect of P and DS rates on δ15N: On the SCLS, high
rates of P decreased δ15N of N2-fixing species, but
mostly at DS0 and the decreases generally peaked at P120
or P160. The negative effect of P on the δ15N of
N2-fixing species indicated the enhancement effect of P on N2
fixation. On the CLS, different effects of DS were observed between N2-fixing
species and the reference crops and DS led to stronger decreases in δ15N
values in reference crops compared with those in N2-fixing species.
Moreover, DS caused higher decreases in δ15N values in bromegrass
than those in alfalfa.
Reducing effect of DS and P treatments on the δ15N values of N2-fixing species could be mainly due to low soil N (non-availability of N with high rates of application of DS) and high dependency of these plants on symbiotic N2 fixation (Haynes, 1980; Hogh-Jensen and Kristensen, 1995; Araújo et al., 2007) and external P. Similarly, in a greenhouse study (Heckman and Kluchinski, 1995), soybean grown on tree leaf- or crop residue-amended soil (20 g residue kg-1 soil; C:N ratio of residues ranged between 17 and 75) exhibited temporary N deficiency until nodulation. Also, they reported that nonnodulating plants were severely N deficient and stunted as a consequence of N immobilization. In their study, nodulating soybean plants grown on leaf or crop residue-amended soil were more dependent on symbiotically fixed N and had lower dry matter yields than the control (unamended) soil. When leaves were composted, the problem of N immobilization was avoided and dry matter yield was not reduced. Also, the results of a greenhouse study (Croteau and Zibilske, 1998) on the effect of papermill sludge (at 25 g sludge kg-1 soil) on the growth of snap bean (Phaseolus vulgaris) showed the immobilization of N with sludge application.
De-inking sludge may increase the pH of the soils, due to the high concentrations of Ca and also pH of DS, which is alkaline (Trépanier et al., 1996; Fierro et al., 1999). These factors may be very effective in increasing nodule number, similar to liming, in common bean (Buerkert et al., 1990) or alfalfa (Pijnenberg and Lie, 1990) and may have enhancing effects on N2 fixation (Hansen and Vinther, 2001; Buerkert et al., 1990; Pijnenberg and Lie, 1990).
In the present study, on the SCLS, high rates of P generally decreased δ15N of N2-fixing plants at DS0 only; the decreases generally peaked at P120 or P160. Moreover, application of P and DS may increase percentages of Ndfa and N uptake in forage legumes. Similarly, in a field experiment, McDonagh et al. (1995) found that the application of P and K fertilizers approximately doubled residue yields, with the combined addition of lime more than doubling these yields again, resulting in 88% of Ndfa in the legumes cowpea (Vigna unguiculata) and groundnut (Arachis hypogaea). In their experiment, the highest legume DM yield and N uptake were in the treatments where lime and P and K were applied, where DM yield was almost four times that of the unfertilized control treatment. McDonagh et al. (1995) indicated that the response when lime and P and K were applied was far in excess of the sum of individual responses. This could be attributed to effects of lime on the availability of P for plant uptake and the large response to lime when added together with P was probably due to the alleviation of P deficiency, resulting in a greater demand for Ca.
Suitability of the methods for determination of symbiotic N2 fixation: Lack of inorganic N in soils can increase atmospheric N2 fixation in legumes (Wanek and Arndt, 2002; Haynes, 1980; Hogh-Jensen and Schjoerring, 1994). Present results showed and confirmed the increasing effect of DS on N2 fixation by forage legumes.
The interesting finding of this study is that, despite the fact that estimates of the percentages of Ndfa could not be obtained by the 15N-natural abundance method, the trends in δ15N values nevertheless allowed discrimination among P rates and DS rates (Broadbent et al., 1982; Danso, 1986).
As proposed by McNeill et al. (1996) the natural abundance technique for estimating symbiotic N2 fixation under field conditions is suggested as a potentially useful alternative to the 15N-enrichment method or the TNDM. Based on the results of the present study, the 15N-natural abundance method needs to be evaluated under field conditions and with application of an agronomic amendment such as DS.
CONCLUSIONS
The composition of DS varies considerably due to the origin of the recycled paper and to the making processes. The macronutrient concentrations, i.e., P and N, in de-inking sludge are generally low. Therefore, a potential limitation for the use of fresh DS as a soil amendment is the possibility of N and P deficiencies for adequate plant nutrition and growth.
The addition of P led to significant decreases in δ15N values with both soils. High rates of P generally decreased δ15N of N2-fixing plants. Furthermore, N2 fixation was generally more strongly enhanced by the application of DS than by P addition. In addition, high rates of DS substantially reduced δ15N of forage legumes, due to N deficiency (non-availability of N with high rates of application of DS) and high dependency of these plants on symbiotic N2 fixation and external P.
Finally, the present study revealed that fresh DS is a suitable soil amendment; for N2-fixing legumes in the presence of high rates of DS and without supplemental nitrogen and according to P concentration in DS composition, P120 seems to be an optimum P rate for all DS rates.