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 Similar articles in this journal
Right arrow Similar articles in 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 Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Agricola
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Related Collections
Right arrow Bioremediation and Biodegradation
Right arrow Nutrient Management
Right arrow Water Pollution
Right arrow Ground Water Quality
Journal of Environmental Quality 30:1-10 (2001)
© 2001 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORT
BIOREMEDIATION AND BIODEGRADATION

Bioremediation of Residual Fertilizer Nitrate

I. Laboratory Demonstration of an On-Farm In Situ Pollution Control System

Benjamin U. Ugwuegbu, Shiv O. Prasher and Darakhshan Ahmad

INRS-Santé, Université du Québec, 245 Boulevard Hymus, QC, Canada H9R 1G6

Corresponding author (darakhshan_ahmad{at}inrs-iaf.uquebec.ca)

Received for publication November 30, 1999.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
 CONCLUSION
 REFERENCES
 
This exploratory laboratory study was undertaken to develop and test an in situ bioremediation system intended to point the way toward a possible field application. The proposed method uses a water table management (WTM) system to deliver nutrients or other amendments to subsoil microorganisms for biostimulation and subsequent biodegradation of pollutants in the saturated and unsaturated zones of the soil. The study was carried out on packed soil columns and bioremediation of residual fertilizer nitrate was attempted. Different levels of organic carbon supplement (glucose C) were introduced into these columns via subirrigation in order to supplement the readily available organic carbon levels in the soil. The study was carried out in two experimental setups. The first setup investigated (i) the effect of addition of a high (970 mg L-1) and a low (120 mg L-1) glucose C level and (ii) the efficacy of using the subirrigation system as a method for nutrient delivery in bioremediation of leached nitrate. This setup was monitored with time, depth, and with reference to the nitrate residue in the soil solution. Leached nitrate was denitrified to less than 10 mg L-1 nitrate N at both glucose levels. The second setup investigated the effect of a range of low levels of glucose C on nitrate decontamination, soil pH, and total microbial count in order to find out an optimal glucose C level that reduced the most nitrate and maintained the pH homeostasis of soil.

Abbreviations: OM, organic matter • WTM, water table management


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
 CONCLUSION
 REFERENCES
 
SOILS polluted either from nonpoint sources (agricultural farms) or point sources (industrial activities) lead to ground water contamination by the process of leaching, ultimately contaminating the surface waters that are recharged by ground water. Nonpoint-source pollution now accounts for a larger portion of pollution than point-source pollution (Oberle and Burkart, 1994; Sharpley and Meyer, 1994; Hasler, 1998), and approximately 30 to 50% of the earth's land is affected by nonpoint-source contamination from pesticides, organic manure, and fertilizer N (Pimental, 1993). It is estimated that 20 to 80% of the fertilizer N applied by farmers as nitrate (NO-3) or ammonium (NH+4), which gets converted to nitrate (Van Loosdrecht and Jetten, 1998), is lost through runoff, resulting in unacceptable levels of water pollution (Liaghat and Prasher, 1996). Nitrate is estimated to be the most common agrochemical contaminant of the world's aquifers (Spalding and Exner, 1993; Livingston and Cory, 1998; Groeneveld, et al., 1998; Verstraete and Philips, 1998), especially in humid regions where drain effluent, carrying residual fertilizer nitrate, contributes significantly to pollution (Skaggs et al., 1994). Although nitrate is essential for efficient plant growth, excessive ingestion of water containing more than the stipulated 10 mg L-1 nitrate N limit is harmful to humans and animals (Fletcher, 1991). The increased awareness of the detrimental effects of nitrate pollution has led to the development of water quality management programs for nitrate pollution control (Malik et al., 1994).

Fate of applied fertilizer nitrate has been well studied and documented in relation to its conservation and distribution in the soil profile and root zone, availability to the crop and effect on the crop yield during the cropping season, and its leaching after harvest (Korom, 1992; Zhou et al., 1997; Hasler, 1998). Leaching of residual nitrate through the soil profile is significantly high during winter and early spring (Liang and MacKenzie, 1994). The practice of WTM seems to effectively control leaching of fertilizer nitrate from the root zone during cropping periods and enhances crop yield (Kalita and Kanwar, 1993). Currently, WTM seems to be the best management practice (BMP) for nitrate pollution control during the croping period (Evans et al., 1989, 1990; Madramootoo et al., 1993). Indeed, WTM systems reduce environmental pollution from leaching of fertilizer nitrate by stimulating the natural microbial denitrification process by creating anaerobic conditions (conducive to denitrification) below the root zone during the cropping period. However, nitrate is highly mobile, and the speed of the process may not be sufficient to denitrify efficiently and prevent leaching within a seasonal time frame. Furthermore, since denitrification under WTM depends upon the availability of organic carbon from the surface of the soil, depletion of carbon in the saturated zone will inevitably decrease the rate of denitrification (Drury et al., 1991; Weier et al., 1991).

The major source of nitrate in farm drains originates from residual nitrate remaining in the unsaturated zone at the end of the farming season (Liang et al., 1991). After crop harvest, WTM is not in operation to control or eliminate the nitrate residue in the unsaturated zone, especially below the root zone. Rather, the subirrigation network reverts to a drainage mode that quickly carries any nitrate leached from the unsaturated zone to a receiving surface water body. Also, the period between harvest and the onset of winter is very short and does not allow complete microbial transformation of all the residual nitrate in the unsaturated zone. Therefore, prompt management is necessary to prevent nitrate from leaching during the growing season (Wright et al., 1992), as well as after the harvest and during the nongrowing season over winter.

Nitrate pollution from fertilizer is technically challenging due to the variability in hydrogeologic conditions and agronomic practices and the diffused nature of the pollution (Fletcher, 1991; Skaggs et al., 1994). The objective of this exploratory study was to demonstrate, at the laboratory scale, an inexpensive and easily adaptable nitrate bioremediation technique intended for the development of an on-farm in situ pollution control system. The technology presented will supplement soil organic carbon via subirrigation using a WTM system in order to enhance denitrification. More specifically, the objectives were to (i) test the effectiveness of subirrigation as the method to supply organic carbon to the subsoil and to determine the effect of the organic carbon supplement on the soil nitrate N residue, (ii) follow the soil homeostasis by evaluating changes in soil solution pH and microbial community status as a result of added organic carbon, and (iii) make recommendations on WTM practices in order to sustain agricultural activities in humid regions. These objectives were accomplished by using sandy soils packed in columns and subirrigated with different levels of glucose, used as an example of a type of organic supplement, while monitoring loss of nitrate residue in the soil solution.


   MATERIALS AND METHODS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSIONS
 CONCLUSION
 REFERENCES
 
Field Site and Soil Used
Two sandy soils, excavated from two different sites at the Macdonald Campus farm, St. Amable and St. Benoit (hereafter referred to as Soil 1 and Soil 2, respectively), were used in this study. The topsoil (5 to 10 cm) was scraped off before excavation. Soil characteristics are presented in Table 1. The main difference between the two soils was the organic matter (OM) content (3.5% for Soil 1 and 1.6% for Soil 2) as determined by loss-on-ignition method (Ball, 1964; Davies, 1974) on three samples of each soil.


View this table:
[in this window]
[in a new window]
  		Table 1. Soil characteristics

 
Column Fabrication and Subirrigation of Soil
Eighteen steel columns (100 cm long x 19.8 cm inside diameter), with sampling ports on the side at 40, 60, and 85 cm from the top, were end-capped with steel plates and fitted with a delivery pipe at the lower end. Slits of 2 mm width were made at 25-mm intervals on the delivery pipe to transfer water and nutrients in the fabricated columns. A geotextile filter of 5-µm pore size was placed over the slits of the delivery pipe. A schematic diagram of the packed soil column is given in Fig. 1 . Six columns were packed with Soil 1 and 12 with Soil 2, all to a bulk density of 1.4 g cm-3, similar to that found in the field. After packing the columns, perforated Teflon tubes were installed at sampling ports for taking liquid samples. A 100-cm-long riser was connected with an elbow to the nutrient delivery pipe to supply nutrients and water to the soil column and to maintain the water table at desired depths. The other end of the riser was attached via PVC tubing to a 4-L Marriotte bottle placed 30 cm above the columns, which served as a subirrigation water reservoir. The columns were placed in a well-ventilated room at a temperature of 24°C. Heating lamps were installed 45 cm above each soil column to cause evaporation of water from the surface of the columns, thus inducing an upward water flow in the columns. The saturated hydraulic conductivity of the soil was determined by the constant-head method (Klute, 1965) on each of the soil columns (Table 1).



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 1. Schematic of a packed soil column for nitrate treatment

 
Nitrate Application and Leaching
To simulate a normal farming practice, all the columns received fertilizer (NO3–N) as calcium nitrate [Ca(NO3)2]. About 4.7 g of Ca(NO3)2 was dissolved in 50 mL of water and poured over a fiberglass mesh placed on the surface of the soil column for uniform distribution without any disturbance of the soil surface. The beaker was rinsed with 500 mL of water and poured over the soil surface. Several rainfall events were simulated at different times to allow leaching of nitrate from the upper soil profile to the lower depths. Rainfall events (Smith et al., 1992) included a once-in-25-yr storm that could occur in Montreal in May (127 mm), a once-in-100 yr (103 mm) storm such as the one that occurred in Montreal in July 1987, and a once-in-20-yr storm (36 mm) (Table 2). During rainfall simulation, water delivery to the soil columns, through subirrigation, was interrupted and excess water was allowed to drain off the columns.


View this table:
[in this window]
[in a new window]
  		Table 2. Experimental parameters in the nitrate study

 
Experimental Setup
Two treatments were used: one involving subirrigation with glucose solution (simulating subirrigation water loaded with a known amount of soluble organic carbon), and the other with water only (the control). Each treatment was randomly assigned to three columns. The influent flowed from a separate Marriotte siphon system, via the riser and the slit delivery pipes at the base of the columns, into the soil profile. The Marriotte siphon and the riser maintained water level in the soil columns to the desired depth of 35 cm from the surface. Water, or water supplemented with glucose C, continuously passed into the columns from the Marriotte reservoir, except during simulation of rainfall. The reservoirs were replenished as the contents were depleted.

The study was carried out in two experiments, each with two stages. The experimental parameters, their stages and durations, OM content of soil used, and the amount of nitrate applied to the columns are given in Table 2. The table also shows the soil columns (identified by numbers) used in each stage, the concentration of glucose C in the subirrigation water and average volume introduced in the columns, and the time (in days) and depth of simulated rainfall events. Soil 1 was used in Experiment I and Soil 2 (with a much lower OM content) was used in Experiment II. Soil columns from the first stage, identified by numbers, were carried forward to the next stage without any modification. Fertilizer was applied to the surface of the columns at a rate of 180 kg N ha-1, and simulated rainfall was used to leach the applied nitrate to lower depths into the soil profile. Soil solutions were obtained before, between, and after the leaching events.

Experiment I
Stage 1—Setup with High Glucose C in Subirrigation Water for Nitrate Reduction in the Soil Profile (55-Day Duration)
Three columns with Soil 1, chosen at random, were subjected to subirrigation with 970 mg L-1 glucose C (glucose treatment), while another three columns of same soil were subirrigated with water (control). Subirrigation commenced 6 d after fertilizer application and nitrate was leached with different depths of rainfall, ranging from 6 to 127 mm, at different times (Table 2). The subirrigation process was interrupted during the leaching events. Soil solutions were obtained from the sampling at 40, 60, and 85 cm below the water table before and after the leaching events and analyzed for nitrate. Forty-five days after the commencement of the treatment, the water table was raised and maintained at the 15-cm depth for 10 d. On Day 55, the saturated hydraulic conductivity was measured and the columns were drained completely for Stage 2.

Stage 2—Setup with Low Glucose C in Subirrigation Water for Nitrate Reduction in the Soil Profile (13-Day Duration)
Columns used in Stage 1 were carried over to Stage 2. Nitrate was applied again on the columns at a rate of 180 kg N ha-1, and a 165-mm rainfall was simulated on each column on the same day (Day 1). Treatment columns were subjected to subirrigation after 5 d with 120 mg L-1 glucose C. Only water was applied to the three previous control columns. Soil solutions were sampled at different times for 13 d and analyzed for nitrate residue.

Experiment II
Stages 1 and 2—Setup for Determining Optimum Glucose Level for Nitrate Reduction in the Soil Profile (124-Day Total Duration)
For the two stages of this experiment, Soil 2 (low OM content) was used to minimize the interference of its hydrolysis into readily available organic carbon, and the long-term effect of addition of a range of glucose levels (0, 20, 70, 150, and 300 mg L-1) was investigated. Twelve columns were packed. Rainfall depths ranging from 36 to 103 mm were simulated on different days during the first 55 d of Stage 1 (Table 2). After subjecting the soil to 96 d of subirrigation with 0, 20, 70, 150, and 300 mg L-1 glucose C in Stage 1, all columns were drained. Another 180 kg nitrate N ha-1 was applied to the columns for Stage-2 studies, and the same (0, 20, 70, 150, and 300 mg L-1) glucose C levels were used in subirrigating the soil for another 35 d. Rainfall depths ranging from 11 to 32 mm were simulated on different days (Table 2). Soil solutions were sampled at the 40-cm depth during Stage 2, on Days 2, 5, 6, 8, 12, 16, 18, 19, 29, 31, 33, 34, and 35, for nitrate and pH measurements.

Sampling and Analytical Methods
Soil solutions were collected from below the water table, 9 h before and after every leaching, at depths of 40, 60, and 85 cm. The samples flowed freely into receiving vials via the sampling tubes located on the side of the columns. The first 5 to 10 mL of the effluent was allowed to go to waste, while the next 10 to 15 mL was collected for analysis. During leaching, 20 to 30 mL of drain effluent was collected for analysis.

Nitrate concentrations in the samples were determined with an ion chromatograph (Water 510 HPLC pump and Star-Ion-A 300 Anion Peek column: 100 x 4.60 mm) equipped with a conductivity detector (Water Model 431), all from Millipore Corp. (Milford, MA). Prior to analysis, each sample was filtered through a 0.45-µm pore size membrane filter. Quantification was done by an external standard calibration curve and pH and temperature were determined directly with a Hanna Scientific combined pH–temperature probe (Springfield Scientific, Springfield, OR).

Total Microbial Count
On Day 35 (at the end of Stage 2, Experiment II), columns were drained completely and soil samples were collected from the top (at a depth of 10 cm) and bottom (at a depth of 85 cm) regions of the columns in sterile tubes. One gram of the soil was added to 4 mL of sodium phosphate buffer and vortexed for 75 s, and serial dilutions up to 10-5 were prepared. A total microbial estimate was carried out by plating 100 µL of 10-3 and 10-5 dilutions of soil suspension on TYc (tryptophan–yeast extract–calcium) agar medium plates (Beringer, 1974). Plates were incubated at 29°C and the microbial colonies were counted everyday for 10 d.

Statistical Analysis
The assessment of delivering nutrients using subirrigation for nitrate bioremediation, was made by monitoring the decrease in nitrate levels at different depths in the soil columns. The experimental design was mixed factorial with spatiotemporal repeated measures and one treatment factor. There were three levels of the spatial repetition factor, depth, and a different number of levels of the temporal repetition factor, with time duration depending on the stages. Statistical analysis involved individual pair-wise comparisons between the control and treatment and a t-test (LSD) for variability using GLM (general linear model) procedures (SAS Institute, 1989). More details on the methodology can be found in Ugwuegbu et al. (2000) and Dutilleul (1998). The significance criterion was at the 5% level. In Stages 1 and 2 of Experiment I, one group of three soil columns received the treatment and another three served as a control. In Stages 1 and 2 of Experiment II, the treatment group received four levels of treatment to determine the optimal range of carbon supplement that would not have an adverse effect on the soil solution while facilitating enhanced reduction of nitrate.


   RESULTS AND DISCUSSIONS
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
CONCLUSION
 REFERENCES
 
Nitrate Reduction with Time and Depth
Effect of Subirrigation with a High Glucose Level on Nitrate Reduction (Experiment I, Stage 1, 55-Day Duration)
The trend of nitrate N appearance and disappearance in soil solution with time and depth is presented in Fig. 2 along with eight rainfall events (six light and two heavy). In general, after each rainfall event, as nitrate leached its level in the soil profile quickly increased and then decreased with time in both treated and control columns. However, the decrease was significantly faster in the columns treated with glucose at all depths, especially after heavy rainfall events. For example, at the 40-cm depth, 6 d after the first heavy (127 mm) rainfall event, on Day 30, the levels in treated columns decreased significantly (at 5% level) from 169 ± 102 to 2 ± 1 mg L-1, while the level of nitrate in the control column remained at 141 ± 55 mg L-1 until Day 42 (Fig. 2a). Two days after the subsequent (eighth) heavy (63 mm) rainfall event on Day 43, the levels decreased in the treated columns from 125 to <1 mg L-1, and in the control columns from 312 ± 256 to about 152 mg L-1, decreasing slowly to 1 mg L-1 in another 10 d. At the 60-cm depth, 6 d after the heavy rainfall event (Day 36), the levels in treated and control columns decreased from 151 ± 16 to less than 1 and 174 ± 41 to 139 ± 65 mg L-1, respectively (Fig. 2b). The latter decreased to 14 ± 11 mg L-1 after another 6 d (Day 42). The rapid decrease of nitrate in the treatment columns is thus attributed to the readily available carbon (glucose) that served as the electron donor required for the rapid denitrification. The nitrate load that reached to the 85-cm depth in both treatment and control columns was less than 10 mg L-1, within the allowable limit, and disappeared quickly (Fig. 2c).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Changes in nitrate N concentration with high level of glucose C (970 mg L-1) in the subirrigation water at the (a) 40-cm, (b) 60-cm, and (c) 85-cm depths (Experiment I, Stage 1). The error bars in the figure and all other subsequent graphs show the standard deviation (SD), showing that 68% of the data obtained (assuming a normal distribution) lie within ±1 SD from the mean (Zolman, 1993; SAS Institute, 1989)

 
Although the depths of rainfall simulations were similar in all columns, equal amounts of nitrate were not leached to corresponding depths in the saturated zone, although uniform packing of soil to the same bulk density was expected to permit equal transport of solutes in the columns. For instance, the amount of nitrate N that leached on Day 30 to the 40-cm depth in both the treatment and control columns varied by 72 to 102 mg L-1 (Fig. 2a). The reason could be that the soil was not thoroughly homogenous when the columns were packed (as homogenizing a large amount of soil is difficult), or that the presence of active organisms, burrowing nonsystematically through the soil and creating nonuniformly distributed macropores (Hole, 1981), caused unequal leaching of nitrate to different depths in different columns.

These results suggest that sufficient total nitrate reduction occurred in both the treatment and control soil columns. However, analysis of variance for nitrate N concentration in the soil solution showed that the carbon treatment's main effect was significant at the 5% level from Day 30 onward, after a heavy rainfall event, as nitrate reduction was faster in treatment columns, especially at the 40- and 60-cm depths (Fig. 2a,b). The absence of a distinct difference in nitrate reduction between the treatment and the control soil columns during the first 30 d of the study is explained by the small amount of nitrate leached to all the depths with light rainfalls (Fig. 2a–c). The increased amount of nitrate leached by heavy (eighth and ninth) rainfall events may have placed a high demand on the organic carbon content in the columns. Thus, the treatment columns, supplemented with external carbon, showed a greater loss of nitrate (within a given time lapse) than the control columns.

After 45 d of experimentation, the water table was raised and maintained for 10 d at a depth of 15 cm from the surface in order to create saturated anaerobic conditions conducive to denitrifying activity by microorganisms, and then the columns were drained. The nitrate concentration in the drain water was less than 10 mg L-1 N in all columns, as anticipated. Supplementing the soil with a readily available carbon source, such as glucose, ensured a faster reduction of the residual nitrate, thus decreasing the time necessary to maintain a high water table. Furthermore, augmenting the indigenous organic carbon with readily available glucose prevented depletion of the native organic carbon from the soil.

The rapid drop in nitrate concentration in the treatment columns clearly indicates a need to supplement the soil organic carbon with a readily available source of organic carbon, and that a WTM practice may be used to introduce supplementary nutrients to any part of the saturated zone. However, the amount of organic carbon added must be controlled and monitored in order to avoid the adverse effects of excessive supplements on the soil agricultural properties, such as low negative redox potential and increased metal mobilization, as presented elsewhere (Ugwuegbu et al., 2000). Thus, the organic carbon concentration was decreased from 970 to 120 mg L-1 in Stage 2.

Effect of Subirrigation with a Low Glucose Level on Nitrate Reduction (Experiment 1, Stage 2, 13-Day Duration)
Figure 3 depicts the changes in nitrate levels that leached after a single heavy (165 mm) rainfall event in both the treatment and control columns. At the 40-cm depth, by Day 13, the nitrate levels in treated columns significantly declined (at 5% level) to below 4 ± 2 mg L-1, while in the control columns the level was 79 ± 49 mg L-1 (Fig. 3a). Similarly, on Day 8 at the 60-cm depths, the nitrate levels decreased significantly (at the 5% level) to 2 ± 1 in treated columns and 36 ± 25 mg L-1 in the control (Fig. 3b). The level of nitrate at the 85-cm depth diminished by 98% in 4 d to 1 mg L-1 in the treatment columns and to 16 ± 7 mg L-1 in the control columns during the same period.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 3. Changes in nitrate N concentration with a low level of glucose C (120 mg L-1) in the subirrigation water at the (a) 40-cm, (b) 60-cm, and (c) 85-cm depths (Experiment I, Stage 2). Error bars show SD

 
The level of denitrification in the control in Stage 2 was lower than that of the previous stage. The rate of denitrification, controlled by the organic carbon content in the soil (Weier et al., 1993), has been shown to be slower with previously decomposed soil OM than with the original soil OM because the OM in the soil leaches out (Bremner and Shaw, 1958). This seems to be the case here, since the soil used in this stage (Soil 1) had been subjected to reducing conditions with intermittent leaching by simulated rainfalls for more than 50 d during Stage 1. Therefore, without a soluble organic carbon supplement, the denitrification rate, as in the control in Stage 2, will continuously decrease, and with time will lead to high levels of nitrate in shallow ground water or to surface water bodies through the drain tiles, contributing to the process of eutrophication. Hence, the slow decrease in nitrate level in the control soil columns in Stage 2 of the experiment suggests a rapid decline in the effectiveness of WTM to sustain denitrification in soils. This highlights the fact that WTM, as currently practiced for nitrate reduction, may not provide a long-term solution for denitrification without carbon augmentation.

Optimizing Glucose Level for Nitrate Reduction (Experiment II, Stages 1 and 2, 124-Day Total Duration)
For an effective decontamination of an agricultural soil it is important that any remediation strategy used should not perturb soil properties during the treatment process (Ugwuegbu, 1996; Ugwuegbu et al., 2000) and therefore, a carbon concentration range that will influence nitrate dissipation without necessarily adversely affecting soil features has to be worked out. This optimum glucose C concentration range was determined in Experiment II using a soil (Soil 2) with a lower (1.6%) OM content to minimize the interference of organic carbon from the hydrolysis of soil OM in this long-duration experiment. In order to simulate the long-term effect of repeated glucose supplements, the columns were treated continuously with 0, 20, 70, 150, and 300 mg L-1 glucose C for 96 d after application of nitrate in Stage 1 (Table 2). The columns were then drained and Stage 2 was started with another application of nitrate.

Figure 4 shows the trend in nitrate appearance and disappearance after rainfall simulation, on columns receiving 0, 20, 70, 150, and 300 mg L-1 glucose C, at the 40-cm depth during Stage 2. Throughout the experimental period (35 d), after the third rainfall event (Day 8), the average nitrate N level was above 40 mg L-1 at the 40-cm depth. The 40-cm depth operated at a maximum denitrification potential in each column from Days 9 to 31; denitrification was not limited by nitrate concentration in any of the columns. Therefore, zero-order kinetics, which is independent of the amount of nitrate present, may be assumed since the lowest amount of nitrate was about 40 mg L-1. This nitrate N concentration has been shown to be sufficient to saturate the enzyme system, thus resulting in a reaction rate being determined by carbon availability rather than nitrate level (Paul and Clark, 1989). At the 40-cm depth, the nitrate loss in the control was significantly different (at the 5% level) from the 20 mg L-1 glucose C treatment. By Day 29, after the fourth rainfall simulation on Day 18, the nitrate levels in the control columns decreased from 82 ± 3 to 23 ± 18 mg L-1, while in the 20 mg L-1 glucose C treatment it diminished from 183 ± 3 mg L-1 to a similar level (26 ± 13 mg L-1) within the same period (Fig. 4a,b). The level of nitrate N reduction during Days 18 and 29 at the 40-cm depth showed significant difference between the treatment columns (probability of significance: Pr > F, 0.0001 at {alpha} = 0.05) (Fig. 5) . In the 20 mg L-1 glucose C treatment the reduction of nitrate N (157 mg L-1) was significantly higher from other treatments with glucose C and the control, which had the least reduction (28.33 mg L-1) of nitrate N. The other three glucose C treatments were significantly and insignificantly different as compared with the control and with each other, respectively (Table 3).



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Changes in nitrate N concentration of subirrigation with glucose C levels at (a) 0, (b) 20, (c) 70, (d) 150, and (e) 300 mg L-1 at the 40-cm depths after 96 d of subirrigation (Experiment II, Stage 2). Error bars show SD

 


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 5. Loss of nitrate N at the 40-cm depth between Days 18 and 29 during treatment with different concentrations of glucose C, after 96 d of subirrigation (Experiment II, Stage 2). Error bars show SD

 

View this table:
[in this window]
[in a new window]
  		Table 3. Reduction of nitrate N at the 40-cm depth between Days 18 and 29 during treatment with different concentrations of glucose C, after 96 d of subirrigation (Experiment II, Stage 2)

 
The poor performance of the glucose C levels above 20 mg L-1 could have resulted from excessive dissolution of mineral ions due to increased microbial activity in the presence of excessive carbon, which may be toxic to denitrifying microorganisms or to their activity. In fact, Tate (1995) has reported that Fe and Mn are more soluble at a low pH, and that acidic conditions are toxic to some microbial metabolic activities. It is known that at a high carbon level, microbial activity (fermentation) is increased, in turn leading to organic acid formation. In order to verify this, the pH of the drain effluent during Stage 2 was measured on different days (Table 4). The pH of the drain effluents for the control and 20 mg L-1 glucose C treatment were very similar and remained above 6, while for all other treatments it was lower by one order in 70 mg L-1 and two orders in 150 and 300 mg L-1 glucose C treatments. Thus, the effect of the lower pH may have resulted in the low performance of glucose C treatments higher than 20 mg L-1.


View this table:
[in this window]
[in a new window]
  		Table 4. pH of drain effluent during nitrate leaching (Experiment II, Stage 2)

 
Assessment of Bioplugging: Hydraulic Conductivity and Microbial Population Estimate (Experiment II, Stages 1 and 2, 124-Day Total Duration)
In in situ bioremediation, a high nutrient concentration results in the proliferation of organisms at the injection ports and the soil zone in the vicinity, leading to biofouling and bioplugging (Shouche et al., 1994; Vandevivere and Baveye, 1992). When this occurs, it is accompanied by a decrease in the hydraulic conductivity of the soil. In this study, the initial saturated hydraulic conductivity in Soils 1 and 2 (77.6 ± 33.7 and 142.0 ± 16.6 mm h-1, respectively, Table 1) did not change significantly during the experimental period, finishing at 78.5 ± 42.0 and 144.4 ± 12.0 mm h-1, respectively. A comparison of the estimated biomass density close to the surface and to the bottom of the soil columns, obtained after Experiment II Stage 2, revealed a population distribution of 106 colony forming units g-1 of soil, often found in agricultural soils (Holben et al., 1988), or was less by one order (Table 5), suggesting that bioplugging did not occur at the bottom of the soil columns as a result of subirrigation with glucose solution. The microbial counts at the 85-cm depth in columns receiving 150 and 300 mg L-1 glucose C were an order higher than the rest. However, similar microbial counts were found at the 10- (near surface) and 85-cm depths with the 300 mg L-1 glucose treatment. Thus, the concentrations of glucose C used in this study did alter the pH but did not alter the microbial population density in the soil columns, although it may have altered the microbial diversity. Since accumulation of organic acids under anaerobic conditions can limit microbial metabolism (Tate, 1995) and growth (Semprini et al., 1991), the anoxic and nutrient-limiting conditions in these columns may have prevented bioplugging.


View this table:
[in this window]
[in a new window]
  		Table 5. Microbial population (total count) distribution close to the surface and bottom of soil columns (Experiment II, Stage 2)

 
Evaluation of Supplementing Soil with Organic Carbon (Experiments I and II, 124-Day Duration)
The similarity of results obtained from the columns receiving the glucose C supplement and the control columns may be due to the initial OM content in the soil. This OM level in the control will continue to decrease unless it is replenished and denitrification will concurrently and gradually decrease with time (Fig. 2a,b and Fig. 5). Thus, there is an optimal carbon level that, when exceeded, could result in a decrease in soil denitrification capacity. Figure 5 clearly indicates that the 20 mg L-1 glucose C level in the present study is the optimum among the concentrations used in this study for Soil 2, reducing more nitrate than the control. This represents a difference of about 81 kg ha-1 of nitrate N using the average porosity of 0.4 [volume of soil x porosity x concentration (mg L-1)]. Evans et al. (1990) have cited a nitrate N leachate load of between 3.7 and 32.4 kg ha-1 yr-1 to the surface water, under low- to high-intensity subsurface WTM practices, decreased by 45% (10 kg ha-1 yr-1 nitrate N) under controlled drainage. Thus the average 81 kg ha-1 of nitrate N loss, obtained from supplementing the subirrigation water with 20 mg L-1 glucose C at a depth of 40 cm (Fig. 5), is more than twice the loss (32.4 kg ha-1 nitrate N) to the drain every year (Evans et al., 1990). Thus, this method, if proven in further long-term field-testings, can be easily used for sustainable agricultural practices in humid regions with controlled subsurface drainage–irrigation systems (based on WTM) (Madramootoo et al., 1993) to supply OM and to create a reduced environment conducive to denitrification of residual fertilizer nitrate leaching from the farm to the subsurface water. In this situation, the use of nitrate fertilizer would not lead inevitably to nitrate pollution. A multiyear experimental field-testing is currently under investigation by our group.

Our results clearly demonstrate that a quick reduction of the residual nitrate may be effected in the unsaturated zone by raising the water table close to the surface with subirrigation water and using a readily available carbon supplement. This has importance when controlling nitrate losses during the fall and the spring snow melt (Ugwuegbu et al., 1994). In early spring, the major source of nitrate in field drains is the residual nitrate in the unsaturated zone that could be removed by this system, immediately after harvesting and before winter, so that nitrate levels in the drain outflow in early spring would be less than the 10 mg L-1 nitrate N limit. The use of shallow water tables or controlled drainage–subirrigation (CD–SI) has been criticized since it may increase runoff and consequently encourage the mobility of other applied agrochemicals, including herbicides and pesticides. It is, therefore, preferable to raise the water table for just a short period before winter to permit cleanup of nitrate residues in the unsaturated zone. Quick reversal to drainage mode will then allow infiltrations during rainfall events and thus reduce losses in the runoff. To further reduce the effect of organic carbon addition to the soil, especially at input depths, an alternate feeding of organic carbon solution and water to the soil is suggested since the input zone of the soil had the most reduced state of all treatment levels, the OM content of the soil not withstanding. The latter would assist in minimizing adverse effects on soil (Ugwuegbu, 1996; Hasselblad and Hallin, 1998). Such an alternate feeding strategy would be conducive for denitrification throughout the planting season.

Managing Carbon Supplements in Subirrigation Water
The primary goals of this study were to develop a pollution control system for nitrate residue that is inexpensive and also easily adaptable to existing infrastructure in subirrigated fields, to make management recommendations for adopting the technique in the field, and to minimize costs. Managing carbon augmentation for nitrate residue bioremediation should be determined primarily by the nitrate pollution index of water resources in the area. The amount of carbon added, in a given farm situation, will depend on several factors. First, it will depend on the readily available carbon content in the soil profile, especially within the water table, as organic carbon content is correlated with denitrification potential. For a soil with low OM that has been previously subjected to decomposition and leaching, organic carbon supplementation is recommended. Second, the amount of added C will depend on the water table elevation. A deep water table will need more water and carbon than a higher one because of increased volume. However, since in summer rainfalls rarely cause drainage, addition of carbon will not be necessary. The third factor is the texture and structure of soil. A coarse-textured soil with a high hydraulic conductivity will facilitate efficient mass transfer of nitrate into the ground water. In a recent study under WTM, nitrate N levels ranging from 25 to 55 kg ha-1 were found at the 75-cm depth after harvest, and when the carbon that leached simultaneously was not sufficient, the leached nitrate remained within the water body (Zhou et al., 1997). Under these conditions, subirrigation water is suggested to be supplemented with readily available carbon, especially during the wet season. The final factor is climatic conditions that determine whether the cropping season will be wet (e.g., 800 mm rain yr-1) or dry (e.g., 480 mm rain yr-1). Because more nitrate will leach into the water table in a particularly wet year, the organic carbon in the saturated zone of the soil would need to be augmented throughout the wet season. However, in a dry year most of the nitrate would remain within the unsaturated zone, especially after harvest when the water table is lowered, and it might be necessary to facilitate its removal by denitrification. This would regenerate a seemingly pristine condition. Under current WTM, the precipitation during the nongrowing season after harvest and before the planting season (consecutive fall and spring) leaches the nitrate residue out of the soil profile, and eventually poses a threat to the environment (Liang et al., 1991; Liang and MacKenzie, 1994; Zhou et al., 1997). Therefore, a quick decontamination of the soil profile below the root zone should be undertaken before the onset of sub-zero temperatures.

Considering the above management possibilities, it is recommended that those in regions with a coarse-textured sandy soil and adversely high nitrate pollution index of water resources adopt this technique. The most cost effective strategy would be a one-time treatment of nitrate residue in the soil profile below the root zone immediately after harvest, especially in a particularly dry season. In a wet season, subirrigation can be alternated with water and carbon solution. Assuming that similar results are obtained from field studies, and that the nitrate distribution in Soil 2 is similar to that employed by Zhou et al. (1997), using sugar as source of organic carbon, it will cost about $113 ha-1 to treat (after harvest) a 75-mm soil profile starting at 25 mm from the surface, with a subirrigation network located at 100 cm from the surface. This is rather expensive, and therefore, more research is needed to identify a cheaper source of carbon, such as waste water, rather than using a commodity product such as sugar.


   CONCLUSION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
 CONCLUSION
REFERENCES
 
Soil nitrate was reduced to negligible levels in column studies as a result of addition of glucose as a source of organic carbon, which enhanced denitrification in the soil profile, via subirrigation through a WTM system. If the in situ bioremediation method demonstrated in this laboratory study is applicable to the fields, most of the nitrate would be removed during the growing season and after the harvest. There is a rapid decline in the potential of WTM, as currently practiced, to sustain denitrification in soils whose OM has been repeatedly leached and subjected to microbial decomposition. This highlights the fact that WTM, as currently practiced for nitrate reduction in agricultural fields, is efficient but not sufficient over a long period without readily available organic carbon augmentation. This exploratory research sets a precedent since it attempts to minimize nonpoint-source pollution caused by nitrate with in situ bioremediation of the nitrate residue within agricultural plots by introducing nutrients via subirrigation water. It will offer a technical solution to on-farm nitrate pollution that is inexpensive, easy to adapt, and will not require major changes in the current farm practices in humid regions with WTM in place. Successful delivery of nutrients for the bioremediation of nitrate, within the farm boundaries, will be considered a breakthrough toward nitrate residue control if this novel approach to control fertilizer nitrate is demonstrated in the field.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
 CONCLUSION
 REFERENCES
 
The study was supported in part by an INRS-Santé graduate scholarship to B.U. Ugwuegbu.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSIONS
 CONCLUSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
B. U. Ugwuegbu, S. O. Prasher, D. Ahmad, and P. Dutilleul
Bioremediation of Residual Fertilizer Nitrate: II. Soil Redox Potential and Soluble Iron as Indicators of Soil Health During Treatment
J. Environ. Qual., January 1, 2001; 30(1): 11 - 18.
[Abstract] [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 Similar articles in this journal
Right arrow Similar articles in 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 Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Agricola
Right arrow Articles by Ugwuegbu, B. U.
Right arrow Articles by Ahmad, D.
Related Collections
Right arrow Bioremediation and Biodegradation
Right arrow Nutrient Management
Right arrow Water Pollution
Right arrow Ground Water Quality

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