Denitrification potential of soils from constructed and natural wetlands

Denitrification potential of soils from constructed and natural wetlands

Ecological Engineering, 2 (1993) 119-130 Elsevier Science Publishers B.V., Amsterdam 119 Denitrification potential of soils from constructed and nat...

661KB Sizes 0 Downloads 23 Views

Ecological Engineering, 2 (1993) 119-130 Elsevier Science Publishers B.V., Amsterdam

119

Denitrification potential of soils from constructed and natural wetlands P.M. G a l e , I. D6vai, K.R. R e d d y a n d D . A . G r a e t z

Soil and Water Science, University of Florida, Institute of Food and Agricultural Science, Gainesville, FL 32611, USA (Received 5 October 1992; accepted 10 March 1993)

ABSTRACT Laboratory experiments were conducted to determine the NO 3 removal (reduction) potential of wetland soils currently used for the disposal of reclaimed wastewater. Surface soil samples were collected, incubated under anoxic conditions with two levels of NO 3 , and the evolution of N20 was measured with time. Denitrification rates ranged from 0.06 to 0.92 g N m -2 day -1, when 10 mg NO3-N per kg soil was added. Complete reduction of NO 3 to N 2 occurred in the soils collected from two constructed wetlands (mineral soils) and one natural wetland (organic soil). However, the final reduction step was inhibited in the soils collected from two wetlands with organic soils. The inhibition was most likely the result of the lower pH of these soils. The results suggest that the addition of wastewater enhances the potential for denitrification in these soils. INTRODUCTION T h e use o f w e t l a n d s as a m e a n s o f t r e a t i n g w a s t e w a t e r is n o t a n e w c o n c e p t . C o m p i l a t i o n s o f case studies o f v a r i o u s w a s t e w a t e r to w e t l a n d systems have b e e n p r e s e n t e d by R e d d y a n d S m i t h (1987) a n d H a m m e r (1989). A l t h o u g h m o s t o f t h e s e studies h a v e b e e n s h o r t - t e r m investigations, w o r k by D e B u s k a n d R e d d y (1987) a n d K a d l e c a n d Bevis (1990) has s h o w n t h a t N r e m o v a l e f f i c i e n c y in w e t l a n d s d o e s n o t d e c r e a s e with p r o l o n g e d wastewater application. T h e b i o g e o c h e m i c a l p r o c e s s e s involved in t h e loss o f N f r o m f l o o d e d soils a n d s e d i m e n t s h a v e b e e n r e v i e w e d by R e d d y a n d P a t r i c k (1984). T h e f o r m s o f N f o u n d in w a s t e w a t e r i n c l u d e o r g a n i c a l l y b o u n d N, a m m o n i u m

Correspondence to: P.M. Gale, Soil and Water Science, University of Florida, Institute of Food and Agricultural Science, Gainesville, FL 32611, USA. Contribution of the Florida Agric. Exp. Stn. Journal Ser. No. R-02709. 0925-8574/93/$06.00 © 1993 - Elsevier Science Publishers B.V. All rights reserved

120

P.M. G A L E E T AL.

ion and nitrate. Nitrate, upon entering the wetland system, need only diffuse into the anaerobic soil layer in order to be subject to denitrification. A m m o n i u m is subject to both ammonia volatilization and nitrification in the floodwater. Organically bound N can undergo mineralization in the sediment-water column and be lost as described above. Because mineralization, nitrification, and denitrification are microbially mediated processes, conditions conducive to microbial growth will enhance these biochemical processes and promote effective N removal by wetlands. The concept of wastewater application to wetlands evolved from the fact that wetlands combine aerobic and anaerobic environments, which enhances N removal through the previously described processes. Denitrification is the final step in the removal of N from a system. Factors influencing denitrification in wetlands include: (i) presence of denitrifiers, (ii) suitable electron donors (readily available C), (iii) absence of 0 2, (iv) available N oxides to act as terminal electron acceptors, and (v) adequate temperature, moisture, and pH to promote microbial growth (Firestone, 1982; Tiedje, 1982, 1988). Several investigators have demonstrated the dependence of denitrification on soil physico-chemical properties. Recently, Groffman et al. (1991) showed an inverse relationship between soil pH (range = 3.5 to 5.5) and denitrification, and Burford and Bremner (1975) noted that microbially available forms of C are more highly correlated with denitrification than total C. Reddy et al. (1982) showed that denitrification rates are similar when they are normalized with respect to the available C in both mineral and organic soils. The dependence of denitrification on available C suggests that C additions may be useful as a bioremediation technique for NO3-contaminated reservoirs (Smith and Duff, 1988; Obenhuber and Lowrance, 1991). Enhanced removal of N in wetland environments has been the basis for integration of wetlands into wastewater treatment systems. Nitrogen-removal rates in the range of 0.5 to 4 g N m -2 day -1 were measured by Burgoon et al. (1991) in constructed wetlands. In contrast, natural wetlands receiving N-containing wastewater had measured N-removal rates in the range 0.04-0.22 g N m -z day -1 (Richardson and Davis, 1987). In previous experiments (Gale et al., 1993), we reported N-removal rates for constructed and natural wetlands that currently receive inputs of reclaimed water (0.8 x 10 6 1 day-1) containing an average total N concentration of 1.8 mg N 1-1 (with about one third of this N in the form of nitrates). Soil-mediated N removal in these systems was in the range of 0.04 to 0.05 g N m -2 day -1, and we concluded that the majority of N removal in this system was through denitrification. The first objective of the present study was to determine the importance of denitrification as a mechanism for N removal in these wetland soils. Second, we wanted to investigate the

121

DENITRIFICATION POTENTIAL OF WETLAND SOILS

possibility that the denitrification potential was enhanced due to wastewater additions. MATERIALS AND METHODS

Site description The Orange County Eastern Service Area Wastewater Treatment facility, located in central Florida, uses a series of natural and constructed wetlands to further polish the reclaimed wastewater before it enters the surrounding surface water. Wastewater effluent from the treatment facility enters the wetlands through a system of manifolds located along the northern and western boundaries (Fig. 1). Flow is overland across the first constructed wetland (CWl) into the first natural wetland (NWl). Water leaving N W l is collected in a channel. Weirs along the eastern edge of the collection channel regulate the flow of water into the second constructed wetland (CW2). Surface flow from CW2 is discharged into a second natural wetland (NW2) from which it leaves the experimental site. Flow into the system ( = 0.8 × 10 6 1 day -1) has an average N O 3 - N c o n t e n t of 0.6 mg 1-1 A berm encloses and isolates the entire wetland system. A third natural

DiItribu~on

.....

* "-

.......... S .................

'\\

.... ..........-.

,

~, "

-.:.

\\

~

~ ":.. .......

"".

~NW2...."

- ,'; m,-

\\ "\N

.: i ." -~ ~

.:""

'">'"

>k"'-:

./

;:~:2222JJ~::~-J~-~*~u7d£JJJ~-~2J~.~JJJ22222~2222222 v

-:

.......

: ............... ~" .................. -

Nwo~

". ;4< ~.~:~..'"" ......

/

, ...... "

..-"" ~

...j

~

g:::.=~::--

\'X ...... "

................

J. ........

* s~m~ir~ s ~ o n

~ ffg~'~t

J

/

.-~,,~,o.

o I

I FEET

I

Fig. 1. Map of the experimental wetlands system currently receiving inputs of reclaimed wastewater. The figure includes approximate boundaries of individual wetland areas and sampling locations used in current study.

122

P.M. G A L E E T A L .

wetland (NW0) that receives no inputs of nutrients from wastewater is located southwest of the experimental wetlands. Both organic and mineral soils are found in the wetlands at the site. The constructed wetlands (CWl and CW2) are former pine flatwoods areas that were cleared and graded. These areas are currently dominated by herbac e o u s aquatic species. Mineral soils in these wetlands have been classified as Smyrna fine sands (sandy, siliceous, hyperthermic Aeric Haplaquods) with bulk densities of = 1.4 g cm -3. Soils of the natural wetlands (NWl, NW2, and NW0) have a high organic matter content and are classified as Sanibel mucks (sandy, siliceous, hyperthermic Histic Humaquepts). They have bulk densities of --0.1 g cm -3 and woody species dominate the vegetative biomass of these wetlands.

Soil sampling Approximately 5 kg of field moist surface soil (0-15 cm) were collected from three locations in each wetland on 20 November 1990. In the laboratory each soil sample was hand mixed, placed in a polyethylene jar, and stored at 5°C. A subsample of the mixed soil sample was dried at 105°C to a constant weight to determine moisture content. The soil samples were also characterized foi': water-soluble organic C (1 : 1 soil to water extract) by persulfate oxidation and infrared analysis of the CO 2 produced (Oceanographic Institute, Inc.), total C and N by Carlo Erba CNS analyzer (Sturmentazione, Italy), and pH in a 1 : 1 water to soil mixture.

Batch incubation experiments The acetylene ( C 2 H 2) blockage technique (Tiedje, 1982) was used to measure the denitrification potential of the soil samples. The experimental design was a randomized complete block with two levels of NO 3 additions and either N 2 gas only or N 2 plus CEH 2 gas treatments. All treatments were in duplicate, and the entire experiment was repeated. Carbon dioxide (CO 2) evolution was measured as a control. Variation in the CO 2 evolved from the different treatments of the same soil indicates differences in microbial populations due to treatments. We placed 10 g of field moist soil into 50-ml glass vials, 5 ml of an amending solution containing either 2 or 20 ~ g NOa-N m1-1, and swirled to mix. This resulted in addition rates of either 1 or 10/xg N g-1. The vials were sealed with silicone-backed teflon caps and the headspace of the vial was purged with N 2 gas (99.9998%) to remove 0 2. Four replicates of each treatment were included. For half of the replicates, 10 ml of the headspace was replaced with acetylene (C2H2), resulting in a displacement of approximately 25% of the headspace. The samples were then incubated in the dark at 25°C.

123

D E N I T R I F I C A T I O N P O T E N T I A L O F W E T L A N D SOILS

Periodically the concentrations of N20 and C O 2 in the headspace of the sample vials was measured. Carbon dioxide (CO 2) was quantified using a Hewlett-Packard (model 5840) gas chromatograph (gc) equipped with a thermal conductivity detector (TCD). Poropak Q (80-100 mesh, Supelco, Bellefonte, PA) was the packing in a 1.8-m × 2-mm ID stainless steel column. The carrier gas (H 2) flow rate was 10 ml min -1, the injector and detector temperatures were 100°C, and the column temperature was 90°C (isothermal operation). Working standards consisted of CO 2 diluted in He (Scott Specialty Gases). Nitrous oxide (N20) analysis was performed on a Shimadzu gc (GC-14A) equipped with a 10-mCi 63Ni electron capture detector (ECD). A 1.8-m × 2-mm ID stainless steel column packed with poropak Q (80-100 mesh) was also used in this gc. The injector, column (isothermal), and the detector temperature were 65, 40 and 340°C, respectively. The carrier gas (5% methane in argon) flow rate was 30 ml min -1 Working standards consisted of N 2 0 diluted in N 2 gas (Alltech Assoc. Calibration Gas). At the end of the incubation 10 ml of 2 M KC1 was added to each vial. The samples were shaken for an hour and filtered through Whatman No. 42 filter paper. The filtrates were analyzed for NH~-N and NOa-N by automated colorimetric techniques (APHA, 1989). RESULTS AND DISCUSSION

Physico-chemical differences between the constructed and natural wetland soils are shown in Table 1. In general the constructed wetland soils (CWl and CW2) had lower water contents and lower total C and N than

TABLE 1 Selected physical and chemical properties of wetland soils collected from the experimental site Sample

Water content (gkg -1)

Constructed wetlands CWla a 359 CWlb 185 CW2 245 Natural wetlands NWl 896 NW2 859 NW0 888

Total carbon (gkg -1)

Total nitrogen (gkg -1)

Soluble organic C (mgkg -1)

NH4-N (mg kg-1)

NO3-N (mg kg-1)

pH

73 26 20

4.9 0.9 0.7

94 53 75

2.1 8.8 9.2

81.7 4.2 0.9

6.8 5.8 6.1

445 437 484

19.9 16.4 21

126 147 152

22.1 24.2 31.2

0.2 0 0.3

6.2 5.1 4.6

a Sample designations refer to locations shown in Fig. 1.

124

P.M. G A L E E T AL.

10

CWl b

A--..,.,,e ~

8

o~o

dlwut @,11~

8 4

w

-

-

10

CW2

8

II

"i"

~1

c_~,

~,

Io

, °.

, ?

10

z

NW2 8

4

2

10

: 4

0

20

40

80

80

100

Time, hours Fig. 2. Accumulation of N20 (with and without C2H 2 treatment) during denitrification in soil samples collected from each of the experimental wetlands. Data points are averages of the three field replicates with the exception of the CWlb graph which represents soils collected along the b transect only. Error bars represent + one standard deviation. CW1 and CW2 are constructed wetlands; NW1, NW2, and NW0 are natural wetlands.

125

DENITRIFICATIONPOTENTIALOF WETLANDSOILS

the organic soils (NWl, NW2, and NW0) found in the natural wetlands. Significantly higher total C, total N, and NO~--N measured in the samples from site CWla were attributed to the influx of wastewater effluent at this location (Fig. 1). Because all the effluent added to the wetland system flows through this location first, it is not surprising that an accumulation of nutrients (C and N) has occurred. In the natural wetland soils (NWl, NW2, and NW0), N20 accumulated in both the samples treated with C2H 2 and those not treated (Fig. 2). In contrast, the constructed wetland soils (CWlb and CW2) accumulated little N20 when samples were incubated without C2H 2. Accumulation of N20 in untreated samples is indicative of rapid reduction of NO 3 ~ N20 and a slower final reduction step (N20 ~ N2). Thus, the accumulation of N20 in the natural wetland soils incubated without C2H 2 suggests that the final reduction step is the rate-limiting reaction in these soils. When NOf-N was not limiting, the rate and extent of denitrification (depicted as N20 accumulation in the presence of C2H2 in Fig. 2) tended to be higher in the natural wetland soils (NWl, NW2, and NW0) than the constructed wetland soils (CWl and CW2). Greater variability was observed in data for the natural wetland soils than the constructed wetland soils. Differences in the amount and rate of N20 accumulation were observed in samples collected from the CWl wetland (Fig. 3). The rate and magnitude of denitrification in the CWla sample is far greater than that of the CWlb location. This may be attributed to the influx of wastewater effluent at the CWla location. The effluent (which contains C, N, and P in various

20 T

Io, 15 Z ~

10 o >

% z

• 5

/ /

/ ~ C W 10 ,.A~ A ~ Q without C2H2 • with C2H2 CWlb

0 without C2H2 Z~ with C2H2

/ l

0

•/t

~.~Z~--Z~

20

40

60

80

I

1O0

Time, h o u r s Fig. 3. Comparison of N20 (with and without C2H 2 treatment) evolved during denitrification in soil samples collected from the a and b transects in the first constructed wetland. Error bars represent + one standard deviation.

126

P.M. G A L E E T A L .

forms) is applied to the western edge of the distribution wetland and flows overland from the inlets across this wetland. The water thus passes site C W l a before it reaches C W l b (Fig. 1). Sample C W l a had higher soil concentrations of total C, N, soluble organic C, and NO 3-N when collected than samples from the C W l b location (Table 1). The continual input of NO 3 (average concentration of the water applied to the wetlands was 0.6 mg N 1-1) at this location is conducive to the establishment of a productive microbial population capable of denitrification. Total denitrification at site C W l a was approximately 10 times greater than that measured at site CWlb. These results also suggest that most of the NO 3 removal occurs in the front portion of the first constructed wetland, near the inflow point of the wastewater effluent. Denitrification in this wetland also may be enhanced by water m a n a g e m e n t practices for the wetland system. Flow into the first constructed wetland is not continuous. As such, portions of this wetland are periodically subjected to drying. Groffman and Tiedje (1991) showed that drying and rewetting stimulates denitrification and CO 2 evolution in soils. Acetylene inhibition of N 2 0 reduction has been shown to fail with continued exposure of the sample (Tiedje, 1988). Failure of the C2H 2 inhibition is evident in Figs. 2 and 3 as the reduction in rate that occurs after 40 h. Therefore, the rate of denitrification for each sample was determined by regressing the concentration of N 2 0 (in the presence of C2H 2) with time for the period of net N 2 0 accumulation (area of maxim u m slope in Fig. 2). Maximum denitrification rates (NO 3 not limiting) ranged from 0.92 g N m -2 day -1 for site C W l a to 0.06 g N m -2 day -1 for site NW0. At the lower N O 3 - N addition, similar denitrification rates were observed in all the wetland soil samples, when these data were expressed on a weight basis (data not shown). However, when expressed in terms of area, the natural wetland soils had detectably lower denitrification rates. In a wastewater land application system, flow is through a given area and not through a given weight of soil. As has been previously pointed out (Hsieh and Coultas, 1989), N removal in freshwater wetland systems is best described on an areal basis. This is in contrast to the interpretations of Parsons et al. (1991) who suggest that spatial and temporal variability preclude translation of denitrification data to an areal basis. A 3- to 10-fold increase in the denitrification rate was observed with a 10-fold increase in NO 3 concentration in the natural wetlands. This suggests that denitrification in these natural wetland soils is limited by NO 3-N levels. For the constructed wetland soils (CW1 and CW2) a 10-fold increase in NO 3 resulted in only a 1- to 3-fold increase in denitrification. It is possible that C (an energy source) is limiting in these mineral soils (Beauchamp et al., 1980). The soluble organic C (SOC) concentration of

127

DENITRIFICATION POTENTIAL OF WETLAND SOILS

TABLE 2 Denitrification and CO 2 evolution rates for soil samples collected from the experimental site Sample

Denitrification rate (g N m -2 day -1)

CO 2 evolution rate (g C m -2 day -1)

1 p, g N g -1 added

10/zg N g -1 added

1 / z g N g -1 added

10/zg N g -1 added

Constructed wetlands CWla a 0.817 CWlb 0.083 CW2 0.130

0.917 0.163 0.276

2.00 0.94 1.08

1.89 0.99 1.24

Natural wetlands NWl 0.010 NW2 0.020 NW0 0.015

0.102 0.103 0.070

0.47 1.01 0.58

0.45 0.95 0.64

LSD.05

0.054

0.16

0.16

0.056

a Sample designations refer to locations shown in Fig. 1.

the mineral soils was a third to a half of that measured in the organic soils (Table 1). The rate of CO 2 evolution was not affected by the N additions (Table 2). Variations in CO 2 evolved from samples collected within a given wetland were much greater than the differences among wetlands. The samples incubated without C2H2 had similar CO 2 evolution rates (data not shown). Thus the C2H 2 treatment did not seem to inhibit or enhance microbial respiration in these soils, one potential problem with the use of the acetylene blockage technique (Tiedje, 1988). The rate of CO 2 evolution was highly correlated with the soluble organic C content of these soils (r = 0.8512, significant at P = 0.01). A similar relationship was found by Reddy et al. (1982) for CO 2 evolution during denitrification. The ratio of the N 2 0 / N 2 produced by the system during denitrification is an indication of the extent of the reduction process (Firestone, 1982). Because the complete reduction of nitrate produces N 2, higher ratios of N 2 0 / N 2 indicate inhibition of the final reductive step. In this study, the ratios were 0.13, 0.15, 0.20, and 0.18 for soil samples collected from sites CWla, CWlb, CW2, and N W l , respectively. This suggests that denitrification proceeds to completion in these soils. In contrast, the soils collected from the 'other natural wetlands (NW2 and NW0) had N 2 0 / N 2 ratios around 0.84, suggesting an inhibition of the final reductive step. Tiedje et al. (1981) found that the N 2 0 / N 2 ratio increases as the pH decreases, indicating a sensitivity to pH. In the NW2 and NW0 wetland soils the

128

P.M. G A L E E T AL.

ambient pH (4.6 for NW0 and 5.1 for NW2) is well below the optimum range for denitrification (6.0-8.5, Tiedje, 1988). The higher the N 2 0 / N 2 ratio, the more likely the system is of producing a net efflux of N 2 0 from the system. A correlation between the denitrification rate and soluble organic carbon (SOC) was observed previously by Burford and Bremner (1975) and Reddy et al. (1982). In the current study the relationship between the denitrification rate and measured SOC was significant at P = 0.05 (r 2= 0.626). When the C W l a samples were removed from this data set, denitrification was found to be highly correlated with SOC (r 2 = 0.8102; P -- 0.01). The rate of denitrification was strongly correlated to the rate of CO 2 evolution. The correlation coefficient for this relationship was 0.808, P = 0.01. The overall stoichiometric reaction for denitrification (Reddy and Patrick, 1984) is: 5 ( C H 2 0 ) + 4NO 3 + 4 H + ~ 5CO 2 + 2N 2 + 7 H 2 0 The stoichiometry of this reaction indicates that there are 1.25 atoms of C oxidized for every N atom reduced. Samples collected from site C W l a were the only ones that approached the stoichiometric ratio (2.3 to 3.4). In all other samples, the amount of CO 2 produced was higher than that expected from denitrification alone ( > 5). This suggests that fermentative pathways or other electron acceptors were being used for microbial respiration in these soils. As noted by Beauchamp et al. (1989), discrepancies between actual measurements and the stoichiometric relationship for denitrification are indicative of the complexity of the system. It is interesting to note that when N was not limiting all of the soils had ratios in the range of 2 to 8. However, under NO3-1imiting conditions this ratio increased to 15 to 85, with the natural wetland soils having significantly higher values (59 to 82) than the constructed wetland soils. This could be due to a greater availability of available C, alternative electron acceptors, and fermentative microorganisms in the natural wetland soils. CONCLUSIONS

Wetland soils indigenous to the experimental site are effective in reducing the concentration of applied N. The potential for NO 3 reduction (measured in this study) exceeded the N-removal rates measured in a previous study (Table 3). Complete reduction of NO 3 ~ N 2 was observed in the first three wetland soils receiving the inputs of reclaimed wastewater. Inhibition of the final reduction step (N20 ~ N 2) was observed in the control wetland soil and the wetland located at the end of the system. This inhibition was thought to be related to the lower p H of these samples. High

DENITRIFICATIONPOTENTIALOF WETLANDSOILS

129

TABLE 3 Comparison of first-order rate constants for denitrification measured in the present study and floodwater N removal measured previously (Gale et al., 1993) Sample Constructed wetland CWla a CWlb

CW2 Natural wetland NW1 NW2 NW0

Denitrification (day- 1)

N removal (day - 1)

0.11 0.10 0.26

0.21 0.16

0.85 0.58 0.47

0.09 0.12

a Sample designations refer to locations shown in Fig. 1.

molar ratios of CO2 produced to NO3 reduced suggests that electron acceptors other than NO 3 were available in the system. In general, the organic soils (natural wetlands) had higher denitrification rates than mineral soils (constructed wetlands) when denitrification rate was expressed on a weight basis (Table 3). However, when the data were expressed on an areal basis, the mineral soils had higher denitrification rates. The highest denitrification rates were measured in samples collected from the CWla location suggesting that the continual input of wastewater effluent enhanced the potential for denitrification. The input of NO 3 into the experimental wetlands (based on 3-year averages) was 99 mg NO3-N m -z day-k The potential removal rate, as measured in these experiments, is 270 mg NO3-N m -2 day -1. Because the potential removal rate is approximately three times greater than the input rate the system should function as a net sink for NO3-N. This in fact is being observed at the experimental site where the inflow concentration of NO3-N (0.6 mg N 1-1) is reduced to 0.4 mg NO3-N 1-1 at the CWla location, 0.13 mg NO3-N 1-1 at the CWlb location, and 0.01 mg NO3-N 1-1 or less at all other sampling locations (CDM, 1991). REFERENCES American Public Health Association (APHA), 1989. Standard Methods for the Examination of Water and Wastewater, 17th ed. Am. Public Health Assoc., Washington, DC, 1268 pp. Beauchamp, E.G., C. Gale and J.C. Yeomans, 1980. Organic matter availability for denitrification in soils of different textures and drainage classes. Commun. Soil Sci. Plant Anal., 11: 1221-1233. Beauchamp, E.G., J.T. Trevors and J.W. Paul, 1989. Carbon sources for bacterial denitrification. Adv. Soil Sci., 10: 113-142. Burford, J.R. and J.M. Bremner, 1975. Relationships between the denitrification capacities of soils and total, water-soluble and readily decomposable soil organic matter. Soil Biol. Biochem., 7: 389-394.

130

P.M. G A L E E T A L .

Burgoon, P.S., K.R. Reddy, T.A. DeBusk and B. Koopman, 1991. Vegetated submerged beds with artificial substrates. II: N and P removal. J. Environ. Eng., 117: 408-424. Camp Dresser and McKee (CDM), 1991. Orange County Eastern Service Area. Phase III Experimental Wetlands Exemption System. Third Annual Summary Report. CDM, Maitland, FL. DeBusk. W.F. and K.R. Reddy, 1987. Removal of flood water nitrogen in a cypress swamp receiving primary wastewater effluent. Hydrobiologia, 153: 79-86. Firestone, M.K., 1982. Biological denitrification. In: F.J. Stevenson (Ed.), Nitrogen in Agricultural Soils. Monograph 22, Am. Soc. Agron., Madison, WI, pp. 289-326. Gale, P.M., K.R. Reddy and D.A. Graetz, 1993. Floodwater nitrogen removal by constructed and natural wetlands microcosms. J. Water Environ. Res., 65: 162-169. Groffman, P.M. and J.M. Tiedje, 1991. Relationships between denitrification, CO z production and air-filled porosity in soils of different texture and drainage. Soil Biol. Biochem., 23: 299-302. Groffman, P.M., E.A. Axelrod, J.L. Lemunyon and W.M. Sullivan, 1991. Denitrification in grass and forest vegetated filter strips. J. Environ. Qual., 20: 671-674. Hammer, D.A., 1989. Constructed Wetlands for Wastewater Treatment: Municipal, Industrial and Agricultural. Lewis Publ., Chelsea, MI. Hsieh, Y.P. and C.L. Coultas, 1989. Nitrogen removal from freshwater wetlands: Nitrification-denitrification coupling potential. In: D.A. Hammer (Ed.), Constructed Wetlands for Wastewater Treatment: Municipal, Industrial, and Agricultural. Lewis Publ., Chelsea, MI, pp. 493-500. Kadlec, R.H. and F.B. Bevis, 1990. Wetlands and wastewater: Kinross, MI. Wetlands, 10: 77-92. Obenhuber, D.C. and R. Lowrance, 1991. Reduction of nitrate in aquifer microcosms by carbon additions. J. Environ. Qual., 20: 255-258. Parsons, L.L., R.E. Murray and M.S. Smith, 1991. Soil denitrification dynamics: Spatial and temporal variations of enzyme activity, populations, and nitrogen gas loss. Soil Sci. Soc. Am. J., 55: 90-95. Reddy, K.R. and W.H. Patrick, Jr., 1984. Nitrogen transformations and loss in flooded soils and sediments. CRC Critical Rev. Environ. Control, 13: 273-309. Reddy, K.R. and W.H. Smith, 1987. Aquatic Plants for Water Treatment and Resource Recovery. Magnolia Press, Orlando, FL. Reddy, K.R., P.S.C. Rao and R.E. Jessup, 1982. The effect of carbon mineralization on denitrification kinetics in mineral and organic soils. Soil Sci. Soc. Am. J., 46: 62-68. Richardson, C.J. and J.A. Davis, 1987. Natural and artificial wetland ecosystems: Ecological opportunities and limitations. In: K.R. Reddy and W.H. Smith (Eds.), Aquatic Plants for Water Treatment and Resource Recovery. Magnolia Pubi., Orlando, FL, pp. 819-854. Smith, R.L. and J.H. Duff, 1988. Denitrification in a sand and gravel aquifer. Appl. Environ. Microbiol., 54: 1071-1078. Tiedje, J.M., 1982. Denitrification. In: A.L. Page (Ed.), Methods of Soil Analysis, Part 2. Agronomy, 9: 1011-1026. Tiedje, J.M., 1988. Ecology of denitrification and dissimilatory nitrate reduction to ammonium. In: A.J.B. Zehnder (Ed.), Biology of Anaerobic Microorganisms. John Wiley and Sons, New York, NY, pp. 179-244. Tiedje, J.M., R.B. Firestone, M.K. Firestone, M.R. Betlach, H.F. Kasper and J. Sorensen, 1981. Use of nitrogen-13 in studies of denitrification. In: R.A. Krohn and J.W. Root (Eds.), Recent Developments in Biological and Chemical Research with Short Lived Isotopes. Advances in Chemistry. Am. Chem. Soc. Washington, DC, pp. 295-317.