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


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in 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 ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Agricola
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Related Collections
Right arrow Water Content
Right arrow Other Environmental Contamination
Right arrow Air Pollution
Right arrow Soil History
Right arrow Nitrogen
Published in J. Environ. Qual. 33:852-860 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Atmospheric Pollutants and Trace Gases

Nitrite Formation and Nitrous Oxide Emissions as Affected by Reclaimed Effluent Application

Y. Master*,a, R. J. Laughlinb, R. J. Stevensb and A. Shaviva

a The Faculty of Agricultural Engineering, Technion-IIT, Haifa 32000, Israel
b Department of Agriculture and Rural Development, Agricultural and Environmental Science Division, Newforge Lane, Belfast BT9 5PX, Northern Ireland, UK

* Corresponding author (master{at}tx.technion.ac.il).

Received for publication June 17, 2003.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
The effect of irrigation with reclaimed effluent (RE) (after secondary treatment) on the mechanisms and rates of nitrite formation, N2O emissions, and N mineralization is not well known. Grumosol (Chromoxerert) soil was incubated for 10 to 14 d with fresh water (FW) and RE treated with 15NO3– and 15NH4+ to provide a better insight on N transformations in RE-irrigated soil. Nitrite levels in RE-irrigated soil were one order of magnitude higher than in FW-irrigated soil and ranged between 15 to 30 mg N kg–1 soil. Higher levels of NO2– were observed at a moisture content of 60% than at 70% and 40% w/w. Nitrite levels were also higher when RE was applied to a relatively dry Grumosol (20% w/w) than at subsequent applications of RE to soil at 40% w/w. Isotopic labeling indicated that the majority of NO2– was formed via nitrification. The amount of N2O emitted from RE-treated Grumosol was double the amount emitted from FW treatments at 60% w/w. Nitrification was responsible for about 42% of the emissions. The N2O emission from the RE-treated bulk soil (passing a 9.5-mm sieve) was more than double the amount formed in large aggregates (4.76–9.5 mm in diameter). No dinitrogen was detected under the experimental conditions. Results indicate that irrigation with secondary RE stimulates nitrification, which may enhance NO3– leaching losses. This could possibly be a consequence of long-term exposure of the nitrifier population to RE irrigation. Average gross nitrification rate estimates were 11.3 and 15.8 mg N kg–1 soil d–1 for FW- and RE-irrigated bulk soils, respectively. Average gross mineralization rate estimates were about 3 mg N kg–1 soil d–1 for the two water types.

Abbreviations: BOD5, five-day biological oxygen demand • FW, fresh water • RE, reclaimed effluent • WFPS, water-filled pore space


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
UTILIZATION OF RECLAIMED effluent (after secondary treatment) is essential in arid and semiarid regions. One of the potential hazards of RE application is its interference with the nitrogen cycle. Based on N budgets, Shaviv et al. (2001) showed that the N recovery in Mediterranean soils irrigated with RE was 20 to 30% lower than when irrigated with FW. Master et al. (2003) measured gaseous N losses (N2, N2O, and NH3) from a Grumosol (Chromoxerert; United States Soil Conservation Service, 1975) in a short-term (about 50 h) lysimeter and laboratory study and found that the RE enhanced NH3 and N2 emissions but had no significant effect on N2O emissions in the short term. Longer-term effects were not verified.

Relatively high nitrite concentrations (up to 9 mg NO2––N kg–1) were observed in soils irrigated with RE (Master et al., 2003). Nitrite formation at low moisture content implied that nitrification rather than denitrification was its major source. A better assessment of the NO2– source can be studied using labeled nitrogen (Burns et al., 1996). Nitrite accumulation during nitrification may be related to the inhibition of the second nitrification stage (i.e., the oxidation of NO2– to NO3– by Nitrobacter) by high concentrations of free ammonia (Smith et al., 1997a, 1997b). Higher NH3 emissions from RE-irrigated soils (Master et al., 2003) suggest that high concentrations of free NH3 may be responsible for NO2– accumulation in these soils. Enhanced NO2– formation could be also due to presence of organic matter (OM) in the RE or RE-irrigated soils (Stueven et al., 1992). The observed decrease in maize (Zea mays L.) yields in lysimeters irrigated with RE (Shaviv et al., 2001) may have resulted partly due to elevated and toxic (Court et al., 1964) NO2– levels in those soils. Soil moisture status also may affect the extent of NO2– accumulation by determining the dominance and intensity of either NO2––forming process (i.e., nitrification and denitrification). In addition, irrigation of dry soils (which is quite common in arid and semiarid regions) may lead to the sudden removal of environmental limitations and release of nutrients. This can be expressed via large bursts of the products, for example, nitric oxide (Hutchinson et al., 1997); however, it has not been shown for NO2–.

Master et al. (2003) also found evidence for the short-term inhibition of nitrification in RE-irrigated soils. The inhibition was attributed to the O2 stress created by the consumption of the available organic matter in the RE by heterotrophs; an additional reason could be the weakening of the soil nitrifying microbial population due to the prolonged exposure to O2–consuming conditions. The findings of the short-term studies indicated that there is a need for longer-term experiments to examine the factors and mechanisms affecting the above-mentioned transformations.

Smith (1980) demonstrated the role of aggregate size in clayey soils by creating "hot-spots," in which O2 supply is limited and thus denitrification is enhanced. Under RE irrigation such an effect is likely to be accentuated due to additional O2 demand by the effluent. It is thus important to test the effects of aggregate size on gaseous N formation in RE-treated soils.

The three longer-term (10–14 d) laboratory incubations in the present study aimed to provide a deeper insight into (i) the source and factors affecting NO2– formation in RE-irrigated Grumosol; (ii) whether NO2– accumulation occurs only after the initial wetting of the soil by RE; and (iii) the effect of irrigation with RE on the emissions of N2 and N2O and on soil mineralization and nitrification. A preliminary evaluation of the effect of aggregate size in clayey soil on gaseous N emissions was performed as well.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Experiment 1: The Influence of Reclaimed Effluent Quality and Irrigation History on Nitrite Formation
The samples of Grumosol (Table 1) used in this laboratory incubation experiment were collected from agricultural fields in Kibbutz Yagur at Haifa Bay. Samples from the 0- to 20-mm depth were collected from individual plots irrigated with FW (referred to hereafter as "FW soil") and RE (referred to hereafter as "RE soil"). The RE soil had been irrigated with RE of various qualities since 1963 from the nearby Haifa wastewater treatment plant. An area of 100 x 20 m was selected from the two plots. Composite soil samples, collected from at least five locations in each of these areas, were brought back to the laboratory and thoroughly mixed. The soil samples were air-dried to a gravimetric water content (referred to hereafter as "moisture content") of 10% and sieved through a 9.5-mm sieve.


View this table:
[in this window]
[in a new window]
  		Table 1. Physical and chemical properties of the Grumosol soil with fresh water (FW) and reclaimed effluent (RE) irrigation histories used in Experiments 1, 2, and 3.{dagger}

 
The incubations were performed at 30°C. There were five soil–irrigation treatments, eight incubation times, and three replicates. One hundred grams FW soil (on an oven-dry basis) was weighed into each of 24 jars (590 mL each) and the same amount of RE soil weighed into each of 96 jars. Ten milliliters of FW were added to each jar 2 d before the start of the incubation to increase the soil moisture content to 20% (corresponding to water-filled pore space [WFPS] of about 33%) for the purpose of conditioning. At the start of the incubation, each jar received an aliquot of solution containing KNO3 and (NH4)2SO4 to yield a moisture content of 40% (corresponding to WFPS of about 65%) and mineral N concentrations of about 50 mg NO3––N kg–1 and 100 mg NH4+–N kg–1 in each jar (equivalent to application of about 23 kg N ha–1). Solutions applied to FW soil were prepared from FW and those applied to RE soils were prepared from either FW or RE (after secondary treatment, which included initial settling of the solids followed by active aeration at various levels) with a BOD5 of 30, 60, or 100 mg L–1 (where BOD5 is five-day biological oxygen demand). The NH+4 content of the effluents used to prepare the solutions ranged between 35 and 50 mg N L–1. The pH of the applied solutions was 7.5 ± 0.08 in FW treatments and 7.4 ± 0.11 in RE treatments. The lids were loosely fitted on the jars to allow adequate aeration while avoiding excessive moisture escape. Destructive sampling was performed 0, 1, 2, 3, 5, 7, 10, and 14 d following application of aliquots. The soil was extracted with 200 mL of 2.5 M KCl. Jars containing soil KCl slurries were shaken for 1 h in a reciprocating shaker, filtered, and stored at 4°C. Concentrations of NH4+, NO3–, and NO2– were determined using a Lachat (Milwaukee, WI) Quikchem 8000 Autoanalyzer. The analyses were performed within one week after the extraction. The pH was determined in aqueous soil extracts in a separate set of incubations, which were run parallel to the main set.

Experiment 2: The Influence of Moisture Content and Repetitive Reclaimed Effluent Applications on Nitrite Formation
This experiment lasted for 3 wk and was performed only with RE soil (same as in Experiment 1). There were three moisture contents, six incubation times (every week), and three replicates. The soil was treated with RE with a BOD5 of 130 mg L–1. Aliquots of (NH4)2SO4 were added to each jar to bring the soil to a level of about 100 mg NH4+–N kg–1 and to yield moisture contents of 40, 60, and 70% (corresponding to WFPS of about 65, 97, and 100%, respectively). Destructive sampling for the three moisture contents was performed 0, 1, 2, 3, 4, and 6 d following application of aliquots. Here, too, the lids were loosely fitted on the jars during the incubation and mineral N species concentrations were determined in an extraction procedure similar to Experiment 1. The 40 and 70% treatments terminated after 6 d, while the soil at 60% was maintained for two additional weeks, during which it was amended with NH4+–enriched RE solution twice more, 7 and 16 d after the beginning of the experiment. The N content of solutions applied at the second and third amendments was similar to that of the first one. At Day 5 the lids of the jars to be sampled in the second and third weeks were removed. This was done to lower the moisture content to allow for the additional dose of RE. The lids of the jars to be sampled in the third week were removed again on Day 14. Sampling times following second and third fertilizer applications were similar to the times during the first week.

Experiment 3: Nitrite Sources and the Effects of Aggregate Size and Reclaimed Effluent on Dinitrogen and Nitrous Oxide Emissions
Soil and Treatments
This experiment lasted for 10 d and was performed with labeled fertilizer. The Grumosol (Table 1) used in this experiment had been irrigated with RE for the last 5 yr in the Acre Agricultural Experimental Farm (Master et al., 2003). The soil was sieved through a 9.5-mm sieve ("Bulk soil") and divided into two additional fractions: (i) "big aggregates," sieve size between 4.76 to 9.5 mm (about 5% of the composition of the bulk soil), and (ii) "small aggregates," sieve size between 0.5 to 2 mm (about 50% of the composition of the bulk soil).

There were two soil–irrigation treatments, two types of labeling for the bulk soil (15NH4+ or 15NO3–), one type of labeling for the big and small aggregates (only 15NO3–), seven incubation times, and three replicates. Soil weight and incubation conditions were as in Experiment 1. One half of the jars were irrigated with FW and the other half with RE with a BOD5 of 59 mg L–1 and NH4+ content of 36.3 mg L–1. At the start of the incubation, each jar received an aliquot of solution containing NO3– and NH4+ to yield a moisture content of 60% (corresponding to WFPS of about 97%) and mineral N concentrations of about 110 mg NO3––N kg–1 and 120 mg NH4+–N kg–1 in each jar. Nitrate was added as natural abundance KNO3 or as 99 atom % K15NO3 and ammonium was added as natural abundance (NH4)2SO4 or as 99 atom % (15NH4)2SO4. Final enrichments of either NO3–or NH4+ pools after the addition of labeled fertilizer were in the range of 37 to 46 atom % excess. Destructive sampling was performed 0, 1, 2, 3, 4, 6, and 10 d following application of aliquots.

Dinitrogen and Nitrous Oxide Measurement and Soil Analyses
Gas sampling times were similar to the times of destructive sampling, excluding time zero. The lids (fitted with gas sampling ports) of the jars were tightly closed 1.5 h before each sampling time. At other times the lids were loosely fitted on the jars. At the end of the headspace closure period of 3 h, 12-mL headspace samples were transferred to evacuated septum-capped vials (Exetainers; Labco, High Wycombe, UK). The samples were analyzed for the isotopic composition of N2 and the concentration and isotopic composition of N2O by continuous-flow isotope-ratio mass spectrometry (IRMS) (Stevens et al., 1993) within 30 d. The N2O concentration and enrichment were determined using the ratios 45R(45I/44I) and 46R(46I/44I) measured by IRMS (Stevens et al., 1993). The flux of N2O was calculated from the change in N2O concentration with time. The cumulative fluxes were calculated by integration, assuming linear change in flux rates between observation times. The amount of dissolved N2O in the soil solution was calculated using the Bunsen coefficient at 30°C (Tiedje, 1994). After the headspace samples were taken, the soil was immediately extracted and subsequently analyzed for concentrations of the mineral N species as in Experiment 1. The 15N contents of NO3– and NO2– in the extracts were determined using the method of Stevens and Laughlin (1994). The 15N content of NH4+ was determined by diffusion of NH3 into boric acid (Laughlin et al., 1997).

Source of Nitrous Oxide
The procedure of Stevens et al. (1997) was used to calculate the fraction (d) of the N2O flux, which was derived from the labeled NO3– pool. The fraction of N2O flux, which was derived from the NH4+ pool, is therefore (1 – d). Valid calculations of d could only be done whenever the NO3– pool was labeled and there was a significant N2O flux. Hence only values of d for the 15NO3– treatment were used to calculate the cumulative amount of N2O–N derived from the labeled NO3– pool (Table 2). Nitrous oxide can be produced by nitrification (Poth and Focht, 1985), denitrification (Firestone, 1982), dissimilatory NO3– reduction to NH4+ (DNRA) (Paul and Beauchamp, 1989), and chemodenitrification (Chalk and Smith, 1983). DNRA is not thought to be a significant process producing N2O in agricultural soils (Paul and Beauchamp, 1989). Chemodenitrification is a significant source of N2O only whenever soil pH is less than 5 (Chalk and Smith, 1983); therefore, it is not likely to have occurred in our basic soils, which also have a high buffering capacity. It was, however, most likely that denitrification was the most important process producing NO3––derived N2O and that nitrification was the most important process producing NH4+–derived N2O.


View this table:
[in this window]
[in a new window]
  		Table 2. Cumulative amounts of N, as N2O, emitted from various soil fractions during the 10 d following application of fertilizer aliquots and the amount attributable to denitrification for NO3––labeled treatments in Experiment 3.{dagger}

 
Gross Nitrogen Transformation Rates
The gross rates of mineralization and nitrification in Experiment 3 for each time interval were estimated using the sizes and dilutions of the NH4+ and NO3– pools at the beginning and end of the time interval (Barraclough, 1991). Total amounts of N nitrified and mineralized during the study period can be calculated as the sum of the amounts in all time intervals. It was assumed that immobilization and remineralization of 15N-labeled species (which would interfere with the calculations) were not significant during the study period. Mineralization would be favored over immobilization due to the low C to N ratios of the RE (approximately 5) and the soils (approximately 10) (Feigin et al., 1991), and the low levels (less than 2 mg NH4+–N kg–1; e.g., Fig. 1) and 15N content of the NH4+ pool (up to 0.15 atom % excess; Fig. 3a, 3c) after 10 d in NO3––labeled treatments hinted that remineralization of 15N-labeled NO3– was negligible.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 1. Changes in mineral N concentrations in reclaimed effluent (RE) soil irrigated with a five-day biological oxygen demand (BOD5) of 60 mg L–1 in Experiment 1. Data represent means and standard deviations (n = 3) or are smaller than symbols.

 


View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. The enrichments of NH4+, NO3–, NO2–, and N2O during Experiment 3 in the bulk soil irrigated and labeled with (a) fresh water (FW) and 15NO3–, (b) FW and 15NH4+, (c) reclaimed effluent (RE) and 15NO3–, and (d) RE and 15NH4+. Error bars are the standard errors of the means (n = 3) or are smaller than symbols.

 

   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Nitrite Accumulation
Figure 1 shows the change in concentrations of mineral N species in Experiment 1 for RE soil irrigated with a BOD5 of 60 mg L–1. This pattern is typical of nitrification and had been observed in all experiments and treatments. All available NH4+ was consumed, while NO3– followed a parabolic increase (preceded by a 1- to 2-d lag) and a subsequent leveling-off after 5 to 6 d. Nitrite levels usually peaked on Day 2 (Fig. 2) , the maximum concentrations being 11.3 and 28.8 mg NO2––N kg–1 for FW and RE soils, respectively, at 60% (Fig. 2c). Higher levels of NO2– were always formed in soils treated with RE as compared with FW, while in Experiment 1 the amounts of NO2– formed in RE soil were independent of the irrigation water type and quality (Fig. 2a). These observations support the findings of Master et al. (2003) of the short-term stimulation of NO2– formation by RE, but do not provide evidence for the source of NO2–.



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Changes in NO2– concentrations in Experiments (a) 1, (b) 2, and (c) 3. Experiments 1 and 3 were performed at moisture contents of 40 and 60%, respectively. Nitrite did not occur in Experiments 1 and 3 after Day 6. Error bars are the standard errors of the means (n = 3) in all incubations except in Experiment 3 in the bulk soil where the values are averaged for 15NO3–– and 15NH4+–labeled treatments (n = 6).

 
The enrichment and dilution of the NO3– pool in the 15NH4+– and 15NO3––labeled treatments, respectively, imply that nitrification occurred, while the dilution of the NH4+ pool in the 15NH4+–labeled treatments implies the occurrence of mineralization in Experiment 3 (Fig. 3) . Similar patterns of nitrification and mineralization were observed in the incubations with the big and small aggregates (data not shown). Nitrite enrichment was generally closer to the enrichment of NH4+ than to that of NO3–, implying that most of it was formed via NH4+ oxidation. This approach allows only an estimation of NO2– source, since a direct quantification of its production or consumption rates requires the labeling of the NO2– pool (Burns et al., 1995, 1996). The lower moisture content (40%) in Experiment 1 mitigated against the development of anaerobic conditions and supports the conclusion that the vast majority of NO2– was formed via oxidation of NH4+.

In the first sampling time in Experiment 3, the size of the NO2– pool was small (Fig. 2c), but the enrichment was about 6 and 15 atom % 15N in 15NO3–– and 15NH4+– labeled treatments, respectively (Fig. 3). In the 15NO3–– labeled treatments, the labeled NO2– could have been formed by rapid reduction of the NO3– pool, whereas in the 15NH4+–labeled treatments it could have been formed by the nitrification of 15NH4+ during the short time interval between the label application and the soil extraction. The degree of enrichment of the NO2– pool implies that nitrification produced more NO2– than reduction. The reason for NO2– enrichment exceeding that of NH4+ in 15NH4+–labeled treatments on Day 3 (Fig. 3b, 3d) could be because the enrichment of NO2– reflects production of 15NO2– from 15NH4+ on Day 2, whereas the enrichment of the NH4+ pool measured at Day 3 is the true value at that instant. Another reason to the inconsistency between pool enrichment is the possible existence of multiple pools (i.e., the 15NH4+ was not uniformly distributed throughout the soil).

Some degree of transient NO2– accumulation following NH4+ addition is a consequence of the nature of nitrification itself (Venterea and Rolston, 2000) and may occur in both FW and RE soils. Other factors specific to RE soils, however, might have induced higher NO2– formation as compared with FW soils. Nitrobacter is known to be more affected by elevated pH values and lowered levels of O2 than the Nitrosomonas (Alleman, 1985). No direct measurements of O2 concentrations were made. The pH values in FW treatments in Experiment 1 (7.2–7.6; Fig. 4) were within the optimum pH range for Nitrobacter (Alleman, 1985). The values in the RE soil (averaged over all treatments), however, were up to 0.4 to 0.5 units higher. This increase in pH (despite the similar NH4+ concentrations in the two soil types) results in doubling the free NH3 concentration in the RE soil solution, which could have inhibited the activity of Nitrobacter but not that of Nitrosomonas (Anthonisen et al., 1976). This is also supported by the higher NH3 losses from RE-irrigated Grumosol lysimeters (Master et al., 2003). The inhibition of the second nitrification step is consistent with the observed time lags of 1 to 2 d in NO3– formation (parallel to NH4+ consumption) observed in nearly all experiments (e.g., Fig. 1). The rate of NH4+ oxidation only needs to be slightly in excess of NO2– oxidation to generate NO2– accumulation (Smith et al., 1997b).



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 4. Changes in pH values in fresh water (FW) and reclaimed effluent (RE) soil during Experiment 1. The values in RE soil are averaged for irrigation with FW; five-day biological oxygen demand (BOD5) = 30, 60, and 100 mg L–1 (n = 4).

 
The type of water used for irrigation did not affect the NO2– accumulation in RE soil (Fig. 2a), implying that the factor affecting NO2– accumulation originates in the soil itself. Oved et al. (2001) found that the predominant population of NH3–oxidizing bacteria changed from Nitrosospira in soils irrigated with FW to Nitrosomonas in soils exposed to irrigation with RE. It may be that the results of Experiment 1 are actually a consequence of that (or other) microbial change inflicted upon the soil by the prolonged exposure to RE irrigation. The high NO2– levels formed after the first RE application were not induced again at subsequent applications (Fig. 2b), despite the similar increase in NH4+ levels (relative to the pre-wetting levels, which were always negligible) induced with each RE application. When some environmental limitations are removed (e.g., wetting of a dry soil), microbial growth and metabolism can be expressed via large bursts of the products (Hutchinson et al., 1997). A subsequent burst will be lower, unless the soils' microbial activity is again severely impeded by desiccation. During Experiment 2, the soil water content was only lowered to a minimum 38% (Fig. 2b), which was high enough not to induce NO2– peaks similar to those subsequent to the first application. Under the semiarid Mediterranean climate, the weekly intervals between subsequent irrigation events may significantly lower the soil moisture content near the soil surface. In the field experiment we are currently performing with a Grumosol soil, the moisture content in the upper 0- to 5-cm soil layer decreases to 11 to 17% after 3 to 4 d following irrigation (data not published), which is even lower than the initial moisture content of soils used in our laboratory experiments (i.e., 20%). In RE soils this is likely to induce NO2– levels similar to those after the first application (i.e., in the range of 14–16 mg NO2––N kg–1) observed during the incubation experiments.

The highest NO2– accumulation was observed at 60% moisture content (Fig. 2b). Nitrification is usually enhanced by moderately high moisture contents, as long as aeration is adequate (Schmidt, 1982). In our case the soils should have been reasonably aerated owing to adequate oxygen diffusion through the thin (about 3 cm) soil layer in the jar. The lowest NO2– peak occurred at 70%, which implied that the overall nitrification rate was slowed down. Additional evidence for the inhibition of nitrification is that only 58 mg NH4+–N kg–1 had been consumed during the first week at 70% moisture content, as compared with 80 and 71 mg NH4+–N kg–1 at 40 and 60% treatments, respectively (data not shown). Burns et al. (1996) observed similar patterns of moisture effect, where the soil NO2– peaks at 50% moisture content were higher than at 40 or 60%.

Dinitrogen and Nitrous Oxide Fluxes
The conditions in Experiment 3 did not induce production of adequate 15N-labeled N2 amounts to be detected by isotope ratio mass spectrometry (fluxes as low as 20 g N2–N ha–1 d–1 would be detected using our setup, which equals to 0.1 mg N kg–1 d–1 in our experiment), although the soil moisture content of 60% created 97% WFPS. Under such conditions, denitrification to form N2 would have been expected (Firestone, 1982). In previous studies with this Grumosol, N2 formation was detected only at saturated conditions (i.e., at 75%) (Master et al., 2003). It appears that under the experimental conditions prevalent in our experiments, N2O is the dominant gaseous N product emitted from the Grumosol.

The N2O fluxes during the incubation period from the bulk soil and from the big and small aggregates are presented in Fig. 5 , with cumulative fluxes being presented in Table 2. Effluent enhanced the N2O losses from the bulk soil and the big aggregates. On average 0.7 and 1.5% from the applied N fertilizer were lost as N2O from FW- and RE-irrigated bulk soil, respectively, mainly during the first 5 d of the incubation. It is commonly reported that substances rich in organic matter enhance the emissions of N2O. When cattle slurry was supplemented with NH4NO3, the loss of N2O was 2.2% compared with 1.2% for NH4NO3 alone (Clayton et al., 1997). Stevens and Laughlin (2001) showed that liquid manure increased the loss of N2O from 1.5 (in control treatments) to 39.1 nmol N g–1 in the 72 h after application. Doubling the loading rates of effluent applied to soil columns resulted in eightfold increase in N2O losses during the following 24 h (Monnett et al., 1995). No difference in N2O losses between FW- and RE-irrigated Grumosol has been observed during short-term studies (Master et al., 2003). About 1.2% from the applied N was lost in both treatments. It could have been that the RE effect was "masked" by the saturated conditions in the short-term studies, while in the present study it became more clearly pronounced due to lower moisture content (60%).



View larger version (33K):
[in this window]
[in a new window]
  		Fig. 5. Nitrous oxide fluxes in Experiment 3 from (a) bulk soil labeled either with NH4+ or NO3– and (b) big and small aggregates labeled with NO3–. Error bars are the standard errors of the means (n = 3) or are smaller than symbols.

 
Production of denitrification-derived N2O in bulk soils irrigated with RE was higher by 43% than in those irrigated with FW (Table 2). This is consistent with the expectation that the organic matter in the RE induces denitrification and thus denitrification-derived N2O. It is widely accepted that denitrification is restricted to anaerobic microsites (Fillery, 1983). Soil denitrification potential is directly related to the extent of the aggregate anaerobic volume (Smith, 1980), which was assumed to be higher in the big aggregates than in the bulk soil. However, Table 2 shows that the N2O source in both cases was nearly similar. Firestone (1982) states that anaerobic zones will only be formed in soils containing water-saturated aggregates with radius larger than 9 mm. The largest aggregate size in our experiment (less than 4.75 mm in radius) was probably not large enough to stimulate a significant increase of the anaerobic zones (as compared with the bulk soil).

Nitrogen Transformations Rates
Nitrification was stimulated in RE soil as compared with FW soil for all soil fractions (Table 3). This contradicts the previous reports of the short-term (1–2 d) nitrification inhibition by RE (Master et al., 2003). Data on direct stimulation of nitrification by RE are not available; however, there are reports of the general proliferation of various microbial groups (including nitrifiers) in soils irrigated with wastewater (e.g., Tam, 1998). This was attributed to the presence of organic matter and nutrients in the wastewater. The nitrifiers' activity was not assessed during the present study, yet the changes in the population structure of ammonia oxidizers in RE soils (Oved et al., 2001) may indicate that the overall soil nitrification potential could have been affected, too. The maximum activity per cell of Nitrosomonas ranges between 0.011 and 0.023 pmol cell–1 h–1, while that of Nitrosospira is reported to be 0.004 pmol cell–1 h–1 in pure culture (Schmidt and Belser, 1994). This could partially explain the enhanced rates in RE soils, since Nitrosospira is the predominant species in FW-irrigated soils and Nitrosomonas in those irrigated with RE (Oved et al., 2001). The per-cell activity data, however, do not provide information of the size of the nitrifying population nor of the extent of its activity under non-ideal environmental conditions (Schmidt and Belser, 1994).


View this table:
[in this window]
[in a new window]
  		Table 3. Estimation of gross amounts of N nitrified and mineralized over 10 d in Experiment 3.{dagger}

 
The stimulation of nitrification in RE soils (Table 3) is consistent with the enhanced amounts of N2O lost due to nitrification (Table 2). The N2O amounts formed by nitrification comprised approximately 0.3 and 0.6% of the nitrified N in FW- and RE-irrigated bulk soil treatments, respectively. These values are in good agreement with those reported by Goodroad and Keeney (1984), who estimated that in relatively wet soils the ratio of N2O–N evolved to N nitrified is in the range of 0.3 to 1.1%. The stimulation of nitrification in RE soils is also consistent with the enhanced NO2– formation in those treatments (Fig. 2c). Mineralization of the organic N in the RE to NH4+ followed by rapid nitrification may (i) render the NH4+ less available to plants and (ii) increase the amounts of the NO3– leached down the root zone to the ground water. The enhanced potential of NO3– losses under RE irrigation is demonstrated in the study of Williamson et al. (1998), in which higher amounts of NO3– (derived mainly from nitrification) leached down from lysimeters irrigated with dairy farm effluent as compared with the control treatments. Shaviv et al. (2001) also showed that the amounts of NO3– leached from RE-irrigated sandy loam lysimeters during the growing season of maize were 25% higher than from those irrigated with FW.

The amount of N mineralized from the organic fraction in the Grumosol during 10 d was around 30 mg N kg–1 for both treatments (Table 3). This soil mineralization potential (about 30% of the applied N) needs to be accounted for when trying to improve the precision of N application rate. The mineralization in RE soils was not higher than in those treated with FW, probably due to the relatively small organic N amounts supplemented by irrigation as compared with the amounts present in the soils (which were nearly equal in the FW and RE soils; Table 1).


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The enhanced NO2– formation in RE-irrigated Grumosol soil observed during a previous short-term study was confirmed during this longer-term study. High NO2– levels (up to 30 mg N kg–1) occurred in soils irrigated with RE, as compared with much lower (up to 11 mg N kg–1) levels in FW treatments. The soil irrigation history with RE, rather than the irrigation water type, was found to be the dominating factor affecting NO2– accumulation in those soils. The highest amounts were formed at a moisture content of 60% when RE was applied to a relatively dry (i.e., about 20%) Grumosol. These levels of soil moisture are likely to prevail in the Mediterranean semiarid climate when the intervals between subsequent irrigation events are long enough. These findings need further verification in a field experiment.

The maximal level of NO2– formation at a soil moisture content of 60% (as compared with 40 and 70%), together with the enrichment data for various mineral N species subsequent to 15N-labeling, suggest that nitrification was the major process responsible for NO2– formation under the experimental conditions. The higher pH of RE-irrigated soils was assumed to drive the NH3 equilibrium concentrations to higher values, resulting in an inhibition of the second nitrification step and subsequent accumulation of NO2–. The enhanced NO2– formation is consistent with both the increase in overall nitrification rates and N2O losses due to nitrification in RE-irrigated soils. It was postulated that nitrification was enhanced as a consequence of the changes in the population of ammonia-oxidizing bacteria in soils exposed to prolonged RE irrigation.

Application of RE increased the emissions of N2O from 0.7% of the applied N fertilizer in FW treatments to 1.5% in RE treatments. Nitrous oxide was the only denitrification product under the experimental conditions prevalent in our experiments. Denitrification was responsible for 36 to 62% of N2O formation and was not affected by the predominant aggregate size. The combination of a significant soil mineralization potential (30 mg N kg–1) and the RE stimulation of the overall nitrification rate may increase the amounts of potentially available NO3– leaching to the ground water. The findings of this research, combined with other known disadvantages of the secondary RE (salinity, sodicity, high content of pathogens, and organic and inorganic pollutants, etc.) require a much more cautious approach in its application.


   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: 799-804. [Full Text]  




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in 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 ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Agricola
Right arrow Articles by Master, Y.
Right arrow Articles by Shaviv, A.
Related Collections
Right arrow Water Content
Right arrow Other Environmental Contamination
Right arrow Air Pollution
Right arrow Soil History
Right arrow Nitrogen

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal