| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a III, Environmental Pollution Control Program, 249 Agricultural Engineering Bldg., The Pennsylvania State University, University Park, PA 16802
b Dep. of Agricultural and Biological Engineering, 233 Agricultural Engineering Bldg., The Pennsylvania State University, University Park, PA 16802
c Pasture Systems & Watershed Management Research Lab, USDA-ARS, Building 3702, Curtin Road, University Park, PA 16802
d Department of Geosciences, 304 Deike Bldg., The Pennsylvania State University, University Park, PA 16802
Corresponding author (rds13{at}psu.edu)
Received for publication June 19, 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: DEA, denitrification enzyme activity
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The possible fates of soil nitrogen include direct uptake by microbes and plants for subsequent incorporation into biological macromolecules (assimilation), microbial denitrification, and loss to ground water. In addition to plant and microbial uptake, soil nitrogen is also strongly influenced by other soil factors such as the presence of oxygen, soil water content, pH, temperature, and organic carbon (Mosier and Schimel, 1993). In an effort to protect surface and ground water from nitrogen intrusion, best management practices such as riparian buffers and constructed wetlands must optimize nitrogen transformation processes that sequester or eliminate nitrogen before discharge to surface waters or ground water aquifers.
Riparian zones are vegetated and frequently forested transitional ecotones between uplands and streams. These areas are not static, but are constantly changing as a result of natural floodplain processes. Flooding episodes can recreate riparian areas by rerouting and degrading the active channel and aggrading and accumulating new banks and active bars (Huggenberger et al., 1998). In addition to flooding episodes, fallen trees and woody debris can reroute active stream channels. These processes create newly scoured meanders and can leave behind relic channels or stream scars within riparian areas, resulting in a heterogeneous floodplain containing a mixture of soil types, unconsolidated gravel deposits, plant and woody debris, and microbiota.
Research on riparian zones has frequently shown reductions in NO-3–N concentrations as ground water and surface water drained from upland agricultural lands and unsewered residential developments through riparian areas to streams. Many studies have demonstrated that riparian zones can be highly effective at NO-3–N removal, with removal efficiencies greater than 80% (Lowrance et al., 1984; Simmons et al., 1992; Pinay et al., 1993; Jordan et al., 1993). In most studies, denitrification is thought to be the major mechanism of NO-3–N removal, although plant and microbial uptake and assimilation, dissimilatory reduction of NO-3–N to ammonium, and the apparent reduction of NO-3–N as a result of ground water and surface water dilution may also be important (Hill, 1996). More recent studies have suggested that denitrification rates, as measured through laboratory experiments, can underestimate NO-3–N removal when compared with ground water well network studies showing NO-3–N removal at the same site. These discrepancies may be a result of the heterogeneous distribution of microbial hotspots and utilizable carbon sources (e.g., buried plant matter and woody debris) that are missed in small (<50 g) soil samples obtained for microcosm studies of denitrification (Groffman et al., 1996; Gold et al., 1998; Jacinthe et al., 1998).
Although the Appalachian Valley and Ridge physiographic province makes up one of the largest portions of the eastern United States and Chesapeake Bay drainage basin, most investigations of riparian zone denitrification in the eastern United States have been conducted in relatively homogeneous coastal plain soils (Hill, 1996; Lowrance et al., 1997). This may be due to the topographic and hydrologic complexity that is characteristic of the Appalachian region, as well as the low percentage of land area in wetlands. However, the linear topographic features and trellis drainage patterns within the Appalachian region generally place wetlands in close proximity to streams and rivers, either as riparian depressions receiving ground water from adjacent uplands or as floodplain wetlands receiving surface water from overbank flooding (Cole et al., 1997). Riparian wetlands, while comprising less than 1% of the land area within the Appalachian region, are thought to provide important water quality functions by serving as sinks for chemicals and nutrients transported by surface or ground water sources, but these water quality functions are not well characterized within the region (Brooks, 1990; Lowrance et al., 1997). A better understanding of NO-3–N removal dynamics in riparian zones within this region may lead to the improvement of strategies and best management practices to improve regional water quality.
The few studies conducted in the Appalachian Valley and Ridge province have observed variable but significant decreases in NO-3–N concentrations in shallow, poorly drained riparian soils near adjacent streams (Schnabel and Stout, 1994; Schnabel et al., 1996). Altman and Parizek (1995) found that ground water dilution was a major contributor to NO-3–N decreases in a valley setting influenced by several perpendicular ground water flow components. These few studies point to the need for further characterization of water quality functions of riparian soils and wetlands located within the Appalachian Valley and Ridge physiographic province, and suggest that the denitrification potential may be site-specific and dependent on factors different from those observed in more uniform hydrogeologic settings. The objectives of this study were to (i) examine and characterize shallow ground water and NO-3–N flow patterns within a forested riparian zone receiving nitrified domestic wastewater effluent, (ii) determine the efficacy of the riparian zone to remove NO-3–N, and (iii) investigate the factors controlling denitrification rates within the riparian soils.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
The southern edge of the final effluent wetlands was directly adjacent to a 30-m-wide forested riparian zone (Fig. 1) . The riparian zone bordered Roaring Run, a second-order headwater stream classified as a high-quality cold water fishery by the Pennsylvania Department of Environmental Protection. The predominant species within the buffer strip was eastern hemlock [Tsuga canadensis (L.) Carr.]. A minor shrub species in the understory was smooth azalea [Rhododendron arborescens (Pursh) Torr.]. There were few herbaceous plants within the buffer strip.
|
|
90 cm depth. A small depressional wetland originated in a relic meander scar in the middle of the riparian zone. An ephemeral stream drained through the depressional wetland and meander scar; we observed surface water flows in this stream on only a few days during the 1998 and 1999 summers. Soils within the meander scar–emphemeral stream area were saturated, with nearly 25 cm of A horizon and a clay–gravel subsurface. A topographically higher area was located between the meander scar and Roaring Run. Soils in this area were usually unsaturated to 60 cm depth, and consisted of a 5- to 8-cm A horizon, a B horizon of unstructured dry sand, and a zone of sandy clay. For consistency in nomenclature, the saturated Andover soils near the constructed wetland system are called the preferential flow, or P area soils; the soils in the meander scar–depressional wetland are MS area soils; and the soils in the topographically elevated area between the depressional wetlands and Roaring Run are the nonpreferential flow, or NP area soils (Fig. 1).
Monitoring Wells
To determine ground water levels and collect water samples, we installed shallow monitoring wells constructed of 5.1-cm PVC pipe with 1-cm holes drilled into the belowground portion of the pipe. Porous landscape fabric covered the pipe to prevent intrusion of soil into the pipe. Holes were dug using a 6.35-cm bucket auger until bedrock was reached (maximum depth = 1.3 m). The perforated, screened pipes were placed in the holes, backfilled with clean sand, and packed with clay around the surface to prevent surface water intrusion. Two rows of wells transected the riparian zone parallel to the stream and final effluent ponds, with Row 1 (Wells 1-1 through 1-5) located between the final effluent wetlands and shallow depressional wetland in the meander scar, and Row 2 (Wells 2-1 through 2-4) located in the topographically elevated area of dry soils between the meander scar and Roaring Run (Fig. 1). A third row of wells (Wells 3-1 through 3-5) transected the meander scar close to Roaring Run. Throughout the summer and fall of 1998, water levels in each well were measured on a weekly to biweekly basis to determine the flow pattern of shallow ground water within the site. During the 1999 summer, we monitored the final effluent wetlands and meander scar wells. Water was evacuated from each well with a hand pump, and then a 60-mL sample was collected using a syringe. Samples were immediately filtered in the field (0.45-µm filter), stored at 4°C, and later analyzed for NO-3–N by ion chromatography (American Public Health Association, 1998).
Denitrification Enzyme Activity
Denitrification enzyme activity (DEA) experiments with riparian soils were performed according to Tiedje (1994). We obtained soil cores in July 1999 from three main areas corresponding to the three rows of monitoring wells installed within the riparian study area: (i) saturated soils between Well 1-4 and the final effluent wetlands (P area), (ii) drier soils from the topographically elevated area between Wells 2-1 and 2-2 (NP area), and (iii) saturated soils in the meander scar directly adjacent to Well 3-2 (MS area) (Fig. 1). Four replicate cores were obtained from the P, NP, and MS areas, and soils from each core were segregated into A, B, and C horizons. Soil slurries were prepared by adding 10 g of field-moist soil from each combination of area and horizon to two 75-mL serum bottles. The first "unamended" bottle received 10 mL of 1 g L-1 chloramphenicol and 10 mL ultrapure water. The second "amended" bottle received 10 mL of a 20 mg L-1 solution of NO-3–N [as Ca(NO3)2] instead of ultrapure water, as in the unamended bottle. All bottles were then flushed repeatedly with O2–free N2 gas and capped. The headspace was injected with 0.1 L L-1 acetone-free acetylene to block the activity of nitrous oxide (N2O) reductase and allow us to measure the N2O as the end product of denitrification. Samples were incubated for 1 to 6 h on a reciprocating shaker at 22°C in the dark, and the sample bottle headspace gas was analyzed for N2O concentration on a gas chromatograph equipped with an electron capture detector. The N2O concentrations were adjusted according to Wilhelm et al. (1977) to account for soluble N2O in the aqueous fraction.
A three-factor analysis of variance was conducted to determine the effect of each of the three main factors—area (P, NP, and MS), horizon (A, B, and C), and amendment (unamended and amended)—or any interactions on denitrification rates. Pairwise comparisons were conducted on any significant effects using Tukey's test. Denitrification rates were not normally distributed, so we transformed the data using a Box–Cox transformation. All denitrification rates were reported as untransformed values to represent rates in meaningful units (Schnabel et al., 1996).
In addition to the experiments listed above, we also investigated the effect of carbon amendment to the NP area C horizon soils. The soil was amended with 10 mL of 1 g L-1 chloramphenicol and 10 mL of a 1 g L-1 glucose solution. We also collected soil from the final effluent wetlands and determined denitrification rates on the unamended soil and soil amended with NO-3–N. To a subset of these samples, we added 5 g moist compost obtained from the Penn State composting facility to determine if a low-cost, readily available source of organic carbon would affect denitrification rates in the final effluent wetlands.
Soil moisture content was determined on a 5-g subsample of field-moist soil from each site and horizon. The field-moist soil was weighed, dried at 105°C to a stable mass, and reweighed. Organic carbon content of each dried subsample was determined by combustion on a Shimadzu (Kyoto, Japan) carbon analyzer. Soil concentrations of NO-3–N were determined by extracting a 2-g subsample with 100 mL of a 2 M KCl solution. Extracts were analyzed colorimetrically on a Technicon (Tarrytown, NY) autoanalyzer using the automated cadmium reduction method (American Public Health Association, 1998).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Most studies of NO-3–N removal in riparian zones have monitored sites in which ground water flowed uniformly from upland edge to receiving waterbody, and the corresponding soil drainage sequence was moderately well drained to poorly or very poorly drained (Lowrance, 1992; Simmons et al., 1992; Hanson et al., 1994; Gold et al., 1998). In contrast, the upland–riparian transition soils in this study were poorly drained colluvial soils draining to a riparian depressional wetland isolated from the receiving stream. This spatial change, unlike more uniform coastal soils, occurs frequently in mountainous areas of the Appalachian Valley and Ridge province, and may affect the location of nutrient reduction within riparian areas.
Nitrate Attenuation in Shallow Ground Water
Nitrate in the shallow ground water flowing from the final effluent wetlands was attenuated between the effluent wetlands and the depressional wetland, even during high flow periods in August. In addition, the spatial pattern of NO-3–N concentrations within the network of monitoring wells indicated that ground water flow and NO-3–N depletion occurred within the poorly drained soils between the effluent wetlands and the depressional wetland. Concentrations of NO-3–N in the final effluent ponds of the treatment wetland system varied during the 1998 summer, ranging from 0.2 to 12.0 mg L-1 (Fig. 3A)
. Concentrations did not exceed 1.0 mg L-1 until mid-July and peaked between 3.3 and 12.0 mg L-1 during the period of highest camp occupancy in August. When summer campers vacated the camp in late August, NO-3–N concentrations declined to less than 1.0 mg L-1. With the exception of Well 1-4, NO-3–N concentrations in Row 1 wells nearest the source of high NO-3–N effluent were always less than 1.2 mg L-1. Concentrations in Well 1-4 reached 9.5 mg L-1 in August, but were lower than the final effluent wetlands, and fluctuated in a manner similar to the effluent wetlands. These results indicate that ground water flow between the effluent wetlands and the depressional wetland occurred within the shallow soil horizons in the Well 1-4 area. In contrast, NO-3–N concentrations in the Row 2 wells between the depressional wetland and the stream were always less than 0.2 mg L-1, further indicating that ground water flowed through the soils of the depressional wetland and that the saturated soils from the effluent wetlands to the depressional wetland were effective at NO-3–N removal.
|
We calculated that there were 250 m2 of riparian wetland with potential for NO-3–N removal within the study area, and estimated that NO-3–N removal rates ranged between 3.2 and 18.1 mg N m-2 h-1 (average = 7.5 mg N m-2 h-1, SE = 2.3, n = 6) in 1998 and 3.3 and 21.9 mg N m-2 h-1 (average = 9.3 mg N m-2 h-1, SE = 2.9, n = 6) in 1999 during July and August when treatment system discharges were highest. Xue et al. (1999) observed similar rates of NO-3–N removal from constructed wetland mesocosms. Their denitrification rates were between 2.0 and 11.8 mg N m-2 h-1 when measured with acetylene inhibition techniques, whereas rates were 
9.3 mg N m-2 h-1 when 15N tracers were used to monitor denitrification. However, N removal rates in their water columns were higher, ranging between 12 and 63 mg N m-2 h-1. The authors attribute the higher water column rates to loss of NO-3–N through infiltration into the sediment.
Denitrification Enzyme Activity
Our observations of NO-3–N removal within the saturated soils of the riparian wetlands were further supported by DEA experiments with riparian soils obtained from the P, MS, and NP areas. Mean DEA within the riparian soils ranged from 0 to 210 µg N kg-1 h-1 (Fig. 4)
. Both area and horizon factors had a significant (p 0.001) effect on DEA, while the amendment factor did not (p = 0.072) (Table 1). The area x horizon and area x amendment interactions were also significant (p 0.001 and p = 0.008, respectively).
|
|
0.001) denitrification rates than B and C horizon soils, while rates for B and C soils were not significantly different (p = 0.075). Whereas A horizon soils from the P and MS areas had significantly higher denitrification rates than their respective B and C horizon soils, denitrification rates across all three horizons of NP soils were typically below detection and not significantly different. The area x amendment interaction was due to the significant effect of NO-3–N amendment on denitrification rates within the MS soils (regardless of horizon), but the P and NP soil denitrification rates were not significantly affected by NO-3–N amendment. Denitrification enzyme activity experiments do not provide direct information about in situ rates of denitrification because of altered experimental conditions. However, DEA has been shown to be strongly related to annual denitrification rates in temperate zone soils, and is useful for among-site comparisons as well as for the distribution of rates within horizons at one site (Groffman and Tiedje, 1989; Groffman et al., 1992; Schnabel et al., 1996). The DEA rates measured in our study are comparable with those measured by Groffman et al. (1992). Their DEA rates were between 680 and 732 µg N kg-1 h-1 in the top 15 cm of very poorly drained riparian soils that received similar concentrations of NO-3–N (8–12 mg L-1) through shallow ground water flow from an unsewered residential area. They also observed a significant correlation between NO-3–N removal and surface DEA (r = 0.73, p < 0.01), as well as between NO-3–N removal and microbial biomass carbon (r = 0.63, p < 0.07). While we did not have sufficient samples to statistically analyze similar variables, the water quality monitoring and DEA data in this study suggest that the shallow A horizon soils within the P and MS areas directly adjacent to the effluent wetlands and within the depressional wetland have a higher potential for NO-3–N removal and DEA than unsaturated soils or deeper horizons.
Soils remained saturated during the summer to within several centimeters of the surface due to the relatively constant input of nitrified wastewater discharged from the treatment wetland system, and had organic carbon contents between 4.6 and 8.8% (Table 2). In contrast, the water table in the NP soil area was typically 40 to 60 cm below the surface during the summer. Denitrification enzyme activity in the saturated C horizon soils of the NP area was low or below detection, despite organic carbon concentrations similar to the upper horizons of the P and MS areas. However, it is likely that carbon within the C horizon was highly refractory due to long-term burial within the alluvial riparian soils, and it did not supply a readily available source of labile carbon for denitrifiers (Addy et al., 1999). In addition, DEA within the C horizon of NP soils did not significantly increase with addition of glucose as a carbon source, suggesting that there was not a bacterial population capable of denitrification within these riparian soils.
|
The final effluent wetlands were excavated into the B horizon of the poorly drained colluvial (P area) soil at the upland–riparian fringe. Denitrification enzyme activity within these unamended soils was low (7.2 µg N kg-1 h-1) relative to the saturated, A horizon soil in the P area, but were similar to rates measured in the B horizon soils of the P and MS areas (Fig. 4). When we added a 50% (by mass) mixture of Penn State University compost to the effluent wetland soil, DEA increased dramatically to 127 µg N kg-1 h-1. These results suggest that denitrification within constructed wetlands could be stimulated by the application of soil amendments that increase the labile carbon content of the shallow wetland soil. However, the ability of carbon amendments to sustain high denitrification rates over the long-term is unknown and warrants further investigation.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. D. Robertson and S. L. Schiff Persistent elevated nitrate in a riparian zone aquifer. J. Environ. Qual., March 1, 2008; 37(2): 669 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. A. Kleinman, M. S. Srinivasan, C. J. Dell, J. P. Schmidt, A. N. Sharpley, and R. B. Bryant Role of Rainfall Intensity and Hydrology in Nutrient Transport via Surface Runoff. J. Environ. Qual., July 1, 2006; 35(4): 1248 - 1259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. W. van Driel, W. D. Robertson, and L. C. Merkley Upflow Reactors for Riparian Zone Denitrification J. Environ. Qual., February 2, 2006; 35(2): 412 - 420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Q. Kellogg, A. J. Gold, P. M. Groffman, K. Addy, M. H. Stolt, and G. Blazejewski In Situ Ground Water Denitrification in Stratified, Permeable Soils Underlying Riparian Wetlands J. Environ. Qual., March 1, 2005; 34(2): 524 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Maitre, A.-C. Cosandey, A. Parriaux, and C. Guenat A Methodology to Estimate the Denitrifying Capacity of a Riparian Wetland J. Environ. Qual., March 1, 2005; 34(2): 707 - 716. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. G. Hunt, T. A. Matheny, and K. C. Stone Denitrification in a Coastal Plain Riparian Zone Contiguous to a Heavily Loaded Swine Wastewater Spray Field J. Environ. Qual., November 1, 2004; 33(6): 2367 - 2374. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Vaillant, F. Monnet, H. Sallanon, A. Coudret, and A. Hitmi Use of Commercial Plant Species in a Hydroponic System to Treat Domestic Wastewaters J. Environ. Qual., March 1, 2004; 33(2): 695 - 702. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 | |||
