JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 9 January 2007
Published in J Environ Qual 36:333-342 (2007)
DOI: 10.2134/jeq2006.0139
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Agricola
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Related Collections
Right arrow Wetlands and Aquatic Processes
Right arrow Ecosystem Restoration
Right arrow Biogeochemical Processes

TECHNICAL REPORTS

Wetlands and Aquatic Processes

Denitrification Potential and Organic Matter as Affected by Vegetation Community, Wetland Age, and Plant Introduction in Created Wetlands

Maria E. Hernandez* and William J. Mitsch

Wilma H. Schiermeier Olentangy River Wetland Research Park, Environmental Science Graduate Program and School of Environment and Natural Resources, The Ohio State Univ., 352 W. Dodridge St., Columbus, OH 43202. M.E. Hernandez, current address: Environmental Biotechnology Unit, Institute of Ecology, Km 2.5 Carretera ant. a Coatepec 351, Congregación el Haya, AP 63, Xalapa 91070, Veracruz, México

* Corresponding author (Elizabeth.hernandez{at}inecol.edu.mx)

Received for publication April 7, 2006.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Denitrification potential (DP) and organic matter (OM) in soils were compared in three different vegetation communities—emergent macrophyte, open water, and forested edge—in two 10-yr-old created riverine wetlands. Organic matter, cold water-extractable organic matter (CWEOM), anaerobic mineralizable carbon (AnMC), and DP varied significantly (P < 0.05) among vegetation communities. The surface (0 to 9 cm) soils in the emergent macrophyte community (EMC) showed highest DP (0.07 ± 0.01 mg N h–1 kg–1), OM (84.90 ± 5.60 g kg–1), CWEOM (1.12 ± 0.20 g kg–1), and AnMC (1.50 ± 0.10 mg C h–1 kg–1). In the deeper layer (9 to 18 cm), DP and CWEOM (0.04 ± 0.01 mg N h–1 kg–1 and 1.13 ± 0.20 g kg–1, respectively) were significantly higher in the open water community (OWC) than in the emergent macrophyte and forested edge communities. Plant introduction did not affect DP or OM content and characteristics. After 10 yr of wetland development, mean DP increased 25-fold in the surface layer (from 0.002 to 0.053 mg N h–1 kg–1); OM content more than doubled to 90.80 ± 19.22 g kg–1, and CWEOM and HWEOM increased 2.5 and 2.7 times respectively from 1993 (prewetland conditions) to 2004. Humic acids were the most abundant form of OM in 2004 and 1993 samples. Significant (P < 0.05) positive relationships between DP and OM, CWEOM, and AnMC were found in the surface layer; in the 9- to 18-cm layer, significant positive relationships were found between DP and CWEOM and AnMC.

Abbreviations: AnMC, anaerobic mineralizable carbon • CWEOM, cold water-extractable organic matter • DP, denitrification potential • EMC, emergent vegetation community • FEC, forested edge community • HWEOM, hot water-extractable organic matter • OM, organic matter • OWC, open water community • W1, Wetland 1 • W2, Wetland 2


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
CREATED wetlands are effective, simple, and economical systems for the reduction of nitrate contamination from agricultural runoff (Bachand and Horne, 2000; Mitsch et al., 2001, 2005a). Nitrate removal from the water column in wetlands is performed by plant uptake and microbial transformation that include immobilization and denitrification. Plant uptake and microbiological immobilization result in temporary storages in the system since most nitrogen will eventually return to the wetland via plant death and decomposition. In contrast, denitrification constitutes a real nitrogen sink because in this process bacteria reduce NO3– to nitrogenous gases (N2, NO, N2O) that are emitted to the atmosphere (Clement et al., 2002).

Heterotrophic denitrification is performed by soil facultative anaerobe bacteria, which under anoxic conditions are able to perform activity by using nitrates as a final electron acceptor. Organic carbon is required as an essential electron donor for heterotrophic denitrification. Organic carbon availability is one of the most important factors that affect denitrification activity in soils (Beauchamp et al., 1989). Quantity and characteristics of organic matter (OM) in soils are influenced by the amount and type of vegetation (Boyer and Groffman, 1996). Several studies have investigated OM availability in agricultural and forest soils (Burford and Bremner, 1975; Bijay-Singh et al., 1988; Boyer and Groffman, 1996), and riparian buffer zones (Hill and Cardaci, 2004; Rotkin-Ellman et al., 2004). Denitrification rates have been estimated for created and restored wetlands (e.g., Poe et al., 2003; Teiter and Mander, 2005; Srivedhin and Gray, 2006; Hernandez and Mitsch, 2007), but little research has been done to investigate the influence of characteristics and availability of OM on denitrification activity in these wetlands.

Creation and restoration of wetlands for nitrogen removal from agricultural landscapes in the Mississippi River basin have been proposed as means to mitigate the hypoxia problem in the Gulf of Mexico (Mitsch et al., 2001, 2005a; Mitsch and Day, 2006). In these created wetlands, high nitrogen removal rates are desirable. Therefore, understanding factors that affect denitrification rates in created wetlands receiving non-point-source pollution is important to optimize their ecological design and performance. In studies performed in microcosms and constructed wetlands treating wastewater, different nitrogen removal efficiencies have been observed when wetlands are planted with different species of macrophytes (Bachand and Horne, 2000; Lin et al., 2002; Hume et al., 2002). These studies have suggested that the differences might be due to the type and quantities of organic carbon that the plant species provide.

In this study we investigated links among vegetation communities, quantity and characteristics of OM, and denitrification potential (DP) in the substrate of created marshes in the Midwest, USA. Our objectives were to compare OM content, fractions of OM, and DP in plots covered with different vegetation communities in two created marshes (one planted and one naturally colonized). Because soil samples were archived when the wetlands were created, we were also able to compare current soil parameters to those of nonhydric soils taken before the wetlands were created.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Site Description
The study was performed at the Olentangy River Wetland Research Park (ORWRP), which is located in the north campus area of The Ohio State University in Columbus, OH, USA. Two 1-ha experimental wetlands were constructed in 1993 on alluvial old-field soils adjacent to the third-order Olentangy River. The primary original soil type at the experimental wetlands was a Ross (Rs) series soil, which is a floodplain alluvial soil that ranges from silt loam to silty clay loam to loam (Mitsch and Wu, 1993). Both wetlands had deeper water sections (approximately 60-cm depth) located in the north, central, and southern positions of the basins. The open water areas of the wetlands were surrounded by emergent plant communities with much shallower depths (0 to 10 cm). Water had been pumped through these wetlands since March 1994 and both wetlands had received the same amount of water and had the same flow patterns since they were created. In a whole-ecosystem experiment that continues to the present, one wetland (W1) was planted with 12 plant species in May 1994 while the second wetland (W2) was naturally colonized by macrophytes (Mitsch et al., 1998, 2005b, 2005c).

Soil Sampling
Four soil cores (10-cm diam. x 18-cm depth) were taken in two patches of the emergent macrophyte community (EMC), in the open water community (OWC), and in the forest edge community (FEC) of each wetland (Fig. 1). Single-species patches of emergent vegetation were dominated by Typha spp. or Schoenoplectus tabernaemontani. The open water zone was colonized by floating and submerged macrophytes such as Lemna spp., Potamogeton pectinatus, and Ceratophylum demersum, and by algal metaphyton. The edge zones were colonized mainly by woody species such as Populus deltoides, Acer rubrum, Salix nigra, and Acer negundo. In the emergent zone, the criterion for choosing a soil sampling patch was the domination of a plant species for at least two consecutive years. To establish sampling plots, maps of vegetation communities in these wetlands were utilized to verify which species were present (Mitsch et al., 2005b). A total of 16 soil samples were taken in each wetland in late autumn 2004. In the emergent vegetation patches, four soil samples were taken in random locations in an area of approximately 2 m2. In the open water and edge zones, soil samples were taken near the inflow (one sample), in the center (two samples), and near the outflow (one sample). A change in consistency of the soil occurred in the samples at approximately 9-cm depth; thus cores were separated in two subsections—0- to 9-cm and 9- to 18-cm depths. Soil cores were placed in zip plastic bags and refrigerated at 4°C. Each core was divided in four parts. One quarter was weighed and dried at 105°C until constant weight to determine bulk density. The other three quarters were mixed together and homogenized by hand; roots, dead plant material, and twigs were separated. Soil moisture, chemical analysis, and incubations were performed with these homogenized samples. Soil moisture, DP, total organic carbon (TOC), and fractions of organic carbon were analyzed in triplicate for each sample.


Figure 1
View larger version (103K):
[in this window]
[in a new window]

  		Fig. 1. Two 1-ha experimental wetlands at the Schiermeier Olentangy River Wetland Research Park (ORWRP) at The Ohio State University, Columbus, OH. Letters represent soil sampling plots in the different vegetation communities. OWC, open water community; FEC, forested edge community; T, Typha spp.; S, Schoenoplectus tabernaemontani. T and S are emergent macrophyte communities (EMC).

 
Denitrification Potential and Mineralized Carbon
Denitrification potential rates were measured within 45 d after sampling, using the acetylene block technique, which inhibits the final conversation of N2O to N2 (Tiedje, 1982). Samples of homogenized fresh soil (approximately 15 g dry wt.) were placed in 1000-mL Mason jars; each jar had a gray butyl septum for gas sampling and a 15-cm sealable vent tube (tygon 2-mm i.d.) attached to the lid. Fifty mL of nitrate (10 mg L–1 of N as KNO3) were added to each sample. The jars were closed and flushed with oxygen-free N2 for 2 min at a flow rate of 8 L min–1 to provide anoxic conditions. While the jars were flushed with N2, the tygon tube was open and submerged in water; when flushing was finished, it was closed with a small clamp and 10% of the volume was replaced by acetone-free acetylene. The slurries were incubated at 20 ± 3°C, and headspace gas was sampled at 0, 6, 12, 24, and 30 h. The jars were shaken by hand approximately every 3 h and before gas sampling. The N2O was analyzed using a gas chromatograph (Shimadzu GC-14-A) fitted with a 2-mL sampling loop, two Porapak-Q 1.8-m columns, and a 63Ni electron capture detector. The instrument used ultra pure nitrogen carrier gas (10 mL min–1), and operation temperatures were 40, 80, and 300°C for column, injector, and detector, respectively. Carbon dioxide was analyzed using a Porapak-Q 1.8-m column and a thermal conductivity detector with ultra pure helium as the carrier gas (25 L min–1) at temperatures of 40, 40, and 200°C for column, injector, and detector, respectively. Total denitrification rates were calculated from the linear portion of N2O produced over the sampling time, and anaerobic mineralized carbon from the linear production of CO2 (Hill and Cardaci, 2004). Gas concentrations measured in the headspace (N2O and CO2) were adjusted for the gas in solution using the Bunsen solubility coefficient (Tiedje, 1982). The nitrate concentration remaining in the slurries was analyzed by the cadmium reduction method using a FIA Lachat AutoAnalyzer. Final concentrations were always >2.0 mg L–1 indicating that nitrate was not limiting during the incubations.

Physical and Chemical Analysis
Soil moisture in the soils was analyzed at 105°C until constant weight. Total OM was analyzed by loss at ignition at 550°C for 1 h according to Nairn (1996) and described in Anderson et al. (2005). Open water samples had slightly alkaline pH; thus to avoid carbonate interferences these samples were pretreated with 10 M HCl until no bubbles were observed, dried at 105°C, and combusted as described before.

Labile and stable OM was determined using the sequential OM extractions, according to Nguyen (2000). Organic matter fractions were sequentially extracted by shaking approximately 2 g (on a dry weight basis) of field moist samples on an end over shaker (120 rpm) with 30 mL of cold (21 to 23°C) distilled water for 18 h, then with 30 mL of hot (80°C) distilled water for 18 h. To quantify stable OM, the remaining residue was subsequently extracted once with 30 mL of a mixture 1:1 of HCl (0.1 M) and HF (0.3 M) for 8 h, then once with 30 mL of sodium pyrophosphate for 24 h, and finally twice with 30 mL of sodium hydroxide (0.5M) for 24 h. Between each step, samples were centrifuged for 15 min at 5000 rpm and the supernatant was filtered (45 µm) and saved for TOC analysis. To establish the time for each extraction of the stable organic fraction, extractions were performed and extracted carbon analyzed several times. For acid and pyrophosphate extractions, it was found that a single extraction was sufficient, but for sodium hydroxide extraction it was necessary to perform the extraction twice. Soluble organic carbon in each extract was analyzed with a Shimadzu TOC-50A50A analyzer. Humic acid precipitate was redissolved in 200 mL of water at pH 7 and soluble TOC was analyzed. Total organic carbon in the extracts was converted to OM by multiplying it by the Van Bemmelen factor (1.724) to express the results as percentage of OM.

Twelve air-dried archived soil samples taken in 1993 before the wetland basins were flooded (see Nairn, 1996; Anderson et al., 2005) were analyzed. Four 1993 sample locations matched 2004 sample locations in the open water zone of W1, a Typha patch in W2, and a Schoenoplectus tabernaemontani patch in W1. These samples were analyzed for total OM, DP, and fractions of organic carbon. Homogenized 2004 subsamples were also air-dried (2 wk at room temperature) and reanalyzed (OM, DP, cold water-extractable organic matter [CWEOM], hot water-extractable organic matter [HWEOM], and anaerobic mineralizable carbon [AnMC]) to compare with 1993 samples. This was performed to minimize differences for sample treatment. For humic substances extraction, 2004 dried soil samples from Typha spp. patches were analyzed and the results were compared with field moist extractions. Results of humic substances in air-dried samples were not significantly different from humic substances extracted in field-moist samples; thus air-dried 1993 samples were compared with field-moist humic-substance extractions in 2004 samples.

Statistical Analysis
Statistical analyses were performed with SPSS version 11 for Macintosh (SPSS, 2003a) and version 12 for Windows (SPSS, 2003b). Kolmogrov-Smirnov, Lilliefors', and Shapiro-Wilk's tests were used to check normality. The data fit normal distributions. One-way analysis of variance (ANOVA) with Tukey HSD multiple comparison tests were used to detect differences among the vegetation communities, depths, and wetland age. When variance was not homogeneous, Games-Howell multiple comparison tests were used. Relationships between DP and OM fractions were examined with Pearson product moment correlations. A 5% significance level was used to determine differences among treatments.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Denitrification Potential and Organic Matter in 10-Year-Old Wetland Soils
Denitrification potential varied significantly (P < 0.05) among the vegetation communities and depths (Fig. 2). In the 0- to 9-cm layer, the highest DP was observed in the EMC (0.07 ± 0.01 mg N h–1 kg–1), while in the 9- to 18-cm layer, the highest DP was observed in the OWC (0.04 ± 0.01 mg N h–1 kg–1). Denitrification potential in the EMC was approximately four times higher in the upper layer than in the 9- to 18-cm layer. In the OWC, DP was 2.5 times higher in the 9- to 18-cm layer than in the 0- to 9-cm layer. In the FEC, DP was low and was not significantly (P > 0.05) different between the two layers.


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 2. Denitrification potential of different vegetation community soils in created marshes. Values are means (n = 8), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05. OWC, open water community; EMC, emergent macrophyte community; FEC, forested edge communities.

 
Organic matter content in the surface layer was significantly higher in the EMC (84.90 ± 5.60 g kg–1) than in the OWC (62.70 ± 2.90 g kg–1) and FEC (64.30 ± 5.10 g kg–1) (Fig. 3a). In the 9- to 18-cm layer OM was not significantly (P > 0.05) different among communities. Only in the EMC was OM significantly higher (P < 0.05) in the 0- to 9-cm layer than the deeper layer.


Figure 3
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 3. Organic matter (OM) characteristics in the different vegetation communities in created wetlands: (a) organic matter content; (b) cold water-extractable organic matter (CWEOM) content; (c) hot water-extractable organic matter (HWEOM) content; and (d) carbon mineralized under anaerobic incubation (AnMC). Values are means (n = 8), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05. OWC, open water community; EMC, emergent macrophyte community; FEC, forested edge communities.

 
Cold-water extractable organic matter in the 0- to 9-cm layer was eight times higher in the EMC than in the OWC and approximately four times higher than in the FEC (Fig. 3b). On the other hand, CWEOM in the 9- to 18-cm layer was 2.5 times higher in the OWC than in the EMC and seven times higher than in the FEC. Cold water-extractable organic matter in the FEC did not differ between the two depths. Hot water-extractable organic matter was significantly (P < 0.05) higher in the emergent macrophyte and open water surface soils than in the forested edge surface soils; in the deeper layer no differences in HWEOM were observed among vegetation communities (Fig. 3c). The highest AnMC was observed in the 0- to 9-cm layer of emergent macrophytes followed by open water and forested edge (Fig. 3d). Anaerobic mineralizable carbon was significantly higher (P < 0.05) in the 0- to 9-cm than 9- to 18-cm layer in the emergent macrophytes and forested edge, but in the open water, AnMC was similar in both layers.

In the 0- to 9-cm layer DP, OM, CWEOM, and AnMC were significantly higher in the Typha spp. patches than in Schoenoplectus tabernaemontani patches (Table 1). However, in the 9- to 18-cm layer no significant differences were observed in these parameters. Hot water-extractable organic matter was not significantly (P > 0.05) different in any soil layer in the two patches.


View this table:
[in this window]
[in a new window]

  		Table 1. Denitrification potential and organic matter characteristics in soils covered with different emergent macrophyte species. Values are means ± SE (n = 8).

 
Organic matter extracted by HCl/HF was very low in all samples, ranging from 2.2 to 3.1% of OM, and no significant differences were observed either among the vegetation communities or among the layers (Fig. 4). Fulvic acids extracted at pH 7 ranged from 3.9 to 4.2% of OM, and fulvic acids extracted at pH 10 ranged from 2.1 to 3.1% of OM. No significant differences in fulvic acid content were found among the samples or among the layers. Humic acids were the most abundant forms of OM in all vegetation communities and differences in their solubility were observed among the vegetation communities. Humic acids extracted at pH 7 were most abundant in surface soils in open water and the forested edge (62 and 65% of OM), while humic acids extracted at pH 10 were most abundant in emergent macrophyte soils (53% of OM). In the 9- to 18-cm layer, humic acids extracted at pH 7 were also more abundant in open water soils (40% of OM), while in the emergent macrophyte and forested edge soils, humic acids extracted at pH 10 (54 and 53% respectively) were most abundant.


Figure 4
View larger version (48K):
[in this window]
[in a new window]

  		Fig. 4. Humic substances content in soils covered with different vegetation communities in created marshes. Values are means (n = 8), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05. OWC, open water community; EMC, emergent macrophyte community; FEC, forested edge communities.

 
Comparison of Planted and Unplanted Wetlands
Ten years after plants were introduced to one wetland basin (W1), and only allowed to colonize naturally in the other wetland basin (W2), no effects of plant introduction on DP, OM, CWEOM, HWEOM, and AnMC were observed (Table 2). This pattern was observed in both soil layers.


View this table:
[in this window]
[in a new window]

  		Table 2. Effect of plant introduction on denitrification potential and organic matter characteristics in soils of created wetlands. Values are means ± SE (n = 16).

 
Denitrification Potential and Organic Matter in Soils Before and After Wetland Creation
Mean DP was very low in both the 0- to 9-cm layer (0.002 ± 0.001 mg h–1 kg–1) and in the 9- to 18-cm layer (0.001 ± 0.001mg h–1 kg–1), and no significant differences were observed between the two layers in the 1993 soil samples (Fig. 5). In 2004, 10 yr after the wetlands were initially flooded and 11 yr after the 1993 soil sampling, mean DP increased significantly (P < 0.05)—25-fold in the 0- to 9-cm layer (0.053 ± 0.03 mg h–1 kg–1) and 15-fold in the 9- to 18-cm layer (0.015 ± 0.004 mg h–1 kg–1). Mean OM in the alluvial soils before the wetlands creation was not significantly different between the layers (40.11 ± 1.70 g kg–1 in the 0- to 9-cm layer and 33.63 ± 5.56 g kg–1 in the 9- to 18-cm layer) (Fig. 6a). In 2004, mean OM in the 0- to 9-cm horizon increased significantly (P < 0.05), approximately doubling to 90.80 ± 19.22 g kg–1. In the 9- to 18-cm layer, mean OM increased by 38% from 1993 to 2004 (46.93 ± 3.85 g kg–1) but this increase was not significant (P > 0.05). Mean CWEOM in the 0- to 9-cm layer increased significantly from 0.35 ± 0.02 g kg–1 in 1993 to 1.10 ± 0.40 g kg–1 in 2004 (Fig. 6b). In the 9- to 18-cm layer, mean CWESOM increased from 0.2 ± 0.05 g kg–1 in 1993 to 0.5 ± 0.10 g kg–1 in 2004. Mean HWEOM in the 0- to 9-cm layer increased from 0.4 ± 0.03 g kg–1 in 1993 to 1.10 ± 0.04 g kg–1 in 2004 (Fig. 6c). In the 9- to 18-cm layer, mean HWEOM also increased from 0.24 ± 0.02 g kg–1 in 1993 to 0.60 ± 0.02 g kg–1 in 2004. In 1993, mean AnMC was significantly higher in the 9- to 18-cm horizons (0.76 ± 0.10 mg h–1 kg–1) than the 0- to 9-cm horizons (0.42 ± 0.0 mg h–1 kg–1) (Fig. 6d). In 2004, mean AnMC increased to 1.80 ± 0.20 mg h–1 kg–1 in the 0- to 9-cm layer, but no increase was observed in the 9- to 18-cm layer.


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		Fig. 5. Comparison of denitrification potential (DP) in soils before wetland creation (1993) and 10 yr after wetlands were flooded (2004). Values are means (n = 12), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05.

 

Figure 6
View larger version (18K):
[in this window]
[in a new window]

  		Fig. 6. Comparison of soil organic matter characteristics before (1993) and 10 yr after wetlands were flooded (2004). (a) Soil organic matter (OM) content; (b) cold water-extractable organic matter (CWEOM) content; (c) hot water-extractable organic matter (HWEOM) content; and (d) carbon mineralized under anaerobic incubation (AnMC). Values are means (n = 12), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05.

 
Humic acids were a main component of OM in 1993 soils. Humic acids extracted at pH 7 were significantly higher in 1993 soils taken before the wetlands were created than in 2004 samples that were flooded for 10 yr (Fig. 7).


Figure 7
View larger version (42K):
[in this window]
[in a new window]

  		Fig. 7. Comparison of humic substances content in soils before wetland creation (1993) and 10 yr after wetland basins were flooded (2004). Values are means (n = 12), bars represent standard error, and different letters indicate significant differences at levels of {alpha} = 0.05.

 
Relationships between Denitrification Potential and Organic Matter Characteristics
In the 0- to 9-cm layer, we found a significant positive relationship between OM content and DP (Pearson coefficient = 0.82), between CWEOM and DP (Pearson coefficient = 0.81), and between AnMC and DP. We did not observe significant relationships between HWEOM and DP (Pearson coefficient = 0.33). In the 9- to 18-cm layer, significant positive relationships were found only between CWEOM and DP (Pearson coefficient = 0.91) and AnMC and DP (Pearson coefficient = 0.84).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Vegetation Type, Organic Matter Content, and Denitrification Potential
Denitrification requires anoxic conditions, organic carbon as an electron supply, and nitrate as a terminal electron acceptor (Beauchamp et al., 1989). Organic matter accumulation in wetlands is the net result of primary production and decomposition. Decomposition of dead plant material is slowed significantly in wetlands due to anaerobic conditions; this results in the formation of extensive peat deposits (Debusk and Reddy, 1998; Collins and Kuehl, 2000). In this study we observed that DP and the quantity of OM in created wetlands soils varied among different vegetation types. Emergent macrophyte communities had higher OM content and DP than open water zones and forest edge zones. Our results agreed with a previous detailed spatial study of OM in these wetlands, which found the greatest concentrations of OM along the emergent zone and lowest concentrations in the open water zone (Anderson et al., 2005).

The higher content of OM in the emergent macrophyte zones could be due to a combined effect of high productivity and different chemical composition of structural carbohydrates. The detrital organic pool in wetlands consists of residual organic compounds of plant materials (Wetzel, 1992). Decomposition rates depend on the structural composition of plants; emergent macrophytes contain higher percentages of structural carbohydrates such as cellulose and lignin compared with floating and submerged macrophytes in open water zones (Mitsch and Gosselink, 2000). Decomposition of lignin is slow because it is a recalcitrant organic compound that is not easily hydrolyzed and metabolized.

In the emergent vegetation patches, OM was higher in the surface layers than in the subsurface layers. However, this pattern was not observed in the open water or edge zones. This could be due to the different structure of vegetation that was the source of OM in the different zones. In the emergent vegetation zone, soil OM sources were mainly both aboveground and belowground biomass decay. Hernandez et al. (2004) found that approximately 70% of belowground biomass in these wetlands was in the 0- to 9-cm depth layer. Therefore, it was expected that this layer would have higher OM inputs by both aboveground and belowground biomass decomposition. On the other hand, in the open water zone, OM was not different in the two soil depths. In these zones, OM sources were from both deposition of autothonous material (decomposed algal, floating and submerged macrophytes biomass) and allochthonous material (river sediments). Allochthonous material accumulation in these zones diluted the samples since river sediments were low in OM relative to autothonous material.

The edge zones in these wetlands were dry most of the time with occasional flooding during flood pulsing; this condition favors colonization by the woody species. We believe that oxidation of OM in these areas was more rapid than in more frequently flooded marsh zones; therefore accumulation of organic in these edge patches was lower.

We found a significant linear relationship between DP and OM in the upper layer of these created riverine wetland soils similar to the linear relationship between denitrification and OM found in hardwood forest soils in the upper Rhine floodplain (Brettar and Hofle, 2002). In the deeper layer no linear relationships between OM and DP were observed; we believe that this was due to the small variations in those parameters observed in different vegetation patches in this layer.

Comparison of Planted and Unplanted Wetlands
After 10 yr of wetland development, no effect of plant introduction on DP and OM was observed. This may be due to the dynamics of macrophyte productivity over the time in these created wetlands. In 1997 both wetlands showed similar peak aboveground macrophyte productivity; then W2 (the naturally colonized wetland) showed higher macrophytes productivity than W1 (planted) for four straight years (1998–2002). In 2003 and 2004, W1 showed higher macrophyte productivity than W2. The cumulative OM produced by macrophytes in the last 7 yr is now almost the same in both wetlands (Mitsch et al., 2005c). The effect of plant introduction on DP and OM content and availability could have been different in earlier years of wetland development.

Vegetation Type, Organic Matter Availability, and Denitrification Potential
In experiments under laboratory conditions, the quality of organic carbon plays an important role as a limiting factor for denitrification in sediments of constructed wetlands and rivers. Addition of easily usable organic carbon such as soluble carbohydrates and acetate to the sediments caused higher denitrification rates than the addition of complex carbon molecules such as fulvic and humic acids (Pfenning and McMahon, 1996; Kozub and Liehr, 1999; Srivedhin and Gray, 2006). Cold water-extractable organic matter is a good indicator of physical mobility and availability of OM. Zsolnay and Steindl (1991) found that 85% of CWEOM in agricultural soils was mineralized and 15% was refractory. On the other hand, HWEOM is more likely to be an indicator of the material that is potentially bioavailable (Zsolnay and Gorlitz, 1994). Hot water-extractable organic matter from fertilized agricultural soil was composed of carbohydrates and N-containing compounds such as amino acids and amides (Leinweber et al., 1995). In this study we observed a significant linear relationship between CWEOM and DP in both layers. Similarly, significant correlations have been observed between DP and water-soluble C, and DP and AnMC in agricultural soils (Burford and Bremner, 1975; Bijay-Singh et al., 1988). Also, in riparian buffer zones in Ontario, Canada, significant positive relationships between DP and water-soluble C, and DP and AnMC were described by Hill and Cardaci (2004). However, we did not find a linear relationship between HWEOM and DP, which indicated that HWEOM in these created wetlands was not immediately available for anaerobic mineralization.

Cold water-extractable organic matter and DP were higher in surface soil layers in Typha spp. patches than in Schoenoplectus tabernaemontani. This was probably due to the type of parental material that these plants provide to soil OM. Plant material decomposition depends on structural composition of plants. Typha spp. contains approximately 10% of their dry mass as lignin and 40% as cellulose (Debusk and Reddy, 1998), whereas Schoenoplectus sp. contains 35% of their dry mass as lignin (Hume et al., 2002). Lignin is more resistant to biological breakdown than cellulose. These differences in structural carbohydrates and also differences in plant productivity might be the reasons why Typha provides more available OM for denitrification than does Schoenoplectus tabernaemontani. Studies of decomposition of crop residues have revealed that cellulose eventually becomes a good carbon source for soil denitrifiers because there is an interaction between denitrifiers and anaerobic fermentative cellulose decomposers (Beauchamp et al., 1989).

Cold water-extractable organic matter and DP were higher in the 9- to 18-cm layer of the OWC; this indicates that OM in the deeper layer of this zone was more available than in the surface layer. In general, floating aquatic plants contain less lignin than emergent plants; therefore they are more easily degraded (Janssen and Walker, 1999). A decay coefficient (k) of 3.7 yr–1 has been reported for Lemna gibba (Szabo et al., 2000), while for the emergent macrophyte Phragmites austrialis, k is in the range of 1.0 to 2.5 yr–1 (van der Valk et al., 1991). It has been suggested that due to relatively slow breakdown, emergent aquatic plants may be significant as a stable supply of carbon in variable environments. In contrast, litter from floating macrophytes may provide pulses of rapidly utilizable carbon at different times a year. We believe that in the open water zones, allochthonous OM has been accumulated in the deeper layer providing more available OM than OM in the surface layer.

Denitrification fluxes (µg N m–2 h–1) in created/constructed wetlands have been described in other studies (Poe et al., 2003; Srivedhin and Gray, 2006; Teiter and Mander, 2005). In our study we measured DP based on soil weight; thus we cannot compare our results with other previous results in created wetlands. Compared with DP in natural riparian wetlands, DP in these 10-yr-old created wetlands is eight times lower than in riparian wetlands soils in the Virginia Coastal Plain (Pavel et al., 1996) and 18 times lower than in riparian wetland soils (peat, mix forest, and marsh) in Ontario Canada (Hill and Cardaci, 2004). This might be due to higher OM content in the natural riparian wetlands (100 to 360 g kg–1) compared with an average of 85 g kg–1 in the surface soils of these 10-yr-old created wetlands.

Humic Substance Characteristics
Humic substances are a general category of naturally occurring heterogeneous organic substances that can be characterized as being yellow through black in color, refractory, and having high molecular weight (Calace et al., 1999). Humification takes place under intense biological activity. In the first stage of the process, microorganisms decompose the original plant material into simpler compounds; later these compounds serve as components for the formation of humic acids, fulvic acids, and humin (Collins and Kuehl, 2000). In this study we found that humic substances were the major proportion of OM in created wetlands. In upland soils, 70 to 80% of soil OM is composed of humic substances (McGrath, 1987). Humic acids were the most abundant form of OM in all the vegetation patches in our created wetlands. Nagamitsu et al. (2002) found that rice paddy soils have a higher humic/fulvic ratio than upland soils because wetland conditions are more favorable to the formation of humic acids than fulvic acids. We did not find differences in humic acid content in the different vegetation patches but we did find differences in their solubility. Humic substance solubility gives us some information about their structure. For example, in Irish soil grasslands, humic acids extracted at pH 7 showed almost double the aromaticity than did humic acids isolated at pH 10 and 12 (McGrath, 1987); the same pattern was observed in podzols under oak trees and cleared forests (Simpson et al., 1997). Aromaticity is related to the transformed state of OM; the more aromatic the more transformed. Although we did not investigate the composition of humic acids in this study, humic acids in edge and open water zones may be more transformed than in the emergent vegetation zone, based on their solubility. This was not expected since the hydrological conditions in the open water and the forested edge were different. A possible explanation may be the fact that in the open water zones there was accumulation of river sediments and these sediments probably originated from upland soil erosion. Flooded conditions seemed to favor the production of less aromatic (transformed) humic acids. On the other hand, transitional upland soils (forested edges) and upland flood plain soils (before wetland creation) showed more transformed humic acids.

Ecological Implications
This study showed that wetlands creation enhanced the DP of floodplains, which is beneficial for mitigation of high nitrate concentrations in surface waters. We compared DP and OM in air-dried soil samples taken before our wetlands were created and 10 yr after water was introduced. Organic matter content in the room-temperature-stored preflooded soils was 25% lower than the mean OM content found by Nairn (1996) for the same samples. This may be the effect of the 11 yr of sample storage. Unfortunately we do not have a reference of DP measured before the wetlands were created; thus the effect of 11 yr of sample storage on DP is unknown. Studies have shown that denitrification activity of air-dried soils with nitrate amendments increased seven times relative to the fresh samples after 1 wk of storage but after 7 wk of storage DP decreased and was only 1.5 times greater than fresh soil samples (Luo et al., 1996). It has been suggested that changes in denitrification activity in air-dried samples might be due to changes in carbon availability and persistence of reduction enzymes (Bijay-Singh et al., 1988; Luo et al., 1996). We found approximately 25 times more denitrification activity in the surface layer after 10 yr of wetland development. Although this increase might partially be due to a storage effect on the 1993 soil samples, it appears that the potential for these floodplain wetlands to carry out denitrification was considerably higher 10 yr after the wetlands were created compared with when they were constructed.

Our results suggest that vegetation type plays an important role in the quantity and quality of OM supply for denitrification in the 10-yr-old created wetlands soils. Hydrology influences the establishment of different plant communities in wetlands (Mitsch and Gosselink, 2000). In these wetlands the hydrology is mostly controlled by river water pumped from the Olentangy River. Water enters to these wetlands at their north side, flows southwards through the wetland (Fig. 1), and finally returns to Olentangy River. The residence time is between 1.5 to 4 d, depending of the inflow rate. Both wetlands have three deep water (>50 cm depth) sections located in the inflow, middle, and outflow positions of the basins, surrounded by much shallower sections (20 to 30 cm deep) dominated by emergent plants. Hernandez and Mitsch (2006) found that the highest denitrification rates in these wetlands occur in the shallow permanently flooded areas with emergent macrophyte vegetation near the inflow. Therefore, if wetlands with high denitrification rates are desired, their design should include shallow water areas to allow colonization of emergent macrophytes rather than large open water areas. Also, the self design of created wetlands for nitrate removal may be assisted by planting or seeding adequate high-productivity macrophytes that provide a stable supply of available OM for denitrification.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Significant variations in DP and quantity, and availability of OM were found in zones with different vegetation communities in 10-yr-old created wetlands. Emergent macrophytes zones showed high available OM and DP in the surface soil layers. Humic acids were the most abundant form of OM; in these wetlands soils, a significant variation in the solubility of humic acids was observed in the different vegetation communities. Denitrification potential in soils before the wetlands were created was one-twentieth that after 10 yr; this increase seems to be related to the increase in quantity and availability of OM in the created wetlands.


   ACKNOWLEDGMENTS
 
The authors would like to thank to Monica Noon, Amber Hanna, and Kyle Chambers for their lab assistance in conducting this project, and Anne Altor for her comments to improve this manuscript. Support provided by U.S. Department of Agriculture NRI CSREES Award 2003-35102-13518, by the Mexican National Council for Science and Technology (CONACYT) Scholarship # 169000, and by a Payne grant from the Ohio Agricultural Research and Development Center (OARDC) of The Ohio State University. Olentangy River Wetland Research Park publication 07-001.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 





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 PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Agricola
Right arrow Articles by Hernandez, M. E.
Right arrow Articles by Mitsch, W. J.
Related Collections
Right arrow Wetlands and Aquatic Processes
Right arrow Ecosystem Restoration
Right arrow Biogeochemical Processes

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