Journal of Environmental Quality 30:1904-1910 (2001)
© 2001 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
TECHNICAL REPORT
Atmospheric Pollutants and Trace Gases
Hurricane-Loaded Soil
Effects on Nitric Oxide Emissions from Soil
Ross M. Tabachow,
J. Jeffrey Peirce* and
Christian Essiger
Department of Civil and Environmental Engineering, Duke Univ., Durham, NC 27708
* Corresponding author (peirce{at}duke.edu)
Received for publication January 19, 2001.
 |
ABSTRACT
|
|---|
The nitric oxide (NO) flux from eastern North Carolina soils subjected to flooding from hurricanes were studied in laboratory experiments. Three sites along the Neuse River basin in eastern North Carolina that sustained different intensities of flooding in September 1999 from Hurricane Floyd were examined. Hurricane Floyd impacted the Neuse River basin by inducing flooding that damaged and disabled hog (Sus scrofa) lagoons and municipal wastewater treatment plants. Between approximately 53 and 325 million liters (14 and 86 million gallons) of untreated hog waste and between approximately 5.7 and 34.4 billion liters (1.5 and 9.1 billion gallons) of untreated municipal wastewater are projected to have entered the Neuse River basin, increasing the concentrations of nitrogen (N), phosphorus (P), and total solids. Phosphorus and total solids are projected to have increased 3.2 and 199.2 mg/L, respectively. Total N was projected to have increased by 9.8 mg/L, which is posited to have increased the NO flux from flooded soils for months after the hurricane. Nitric oxide emissions from soil can adversely affect ozone levels in the lower troposphere. Minimization of NO flux from soil is advantageous, protecting air quality as well as conserving valued nitrogen fertilizers. Hurricane-loaded soils were found to produce more than 30 times greater NO emissions than nonflooded soils with NO fluxes ranging from 0.1 to 102.5 ng N/(m2 s).
Abbreviations: NO, nitric oxide • WFPS, water-filled pore space • WWTP, wastewater treatment plant
|
INTRODUCTION
|
|---|
HURRICANE Floyd struck North Carolina on 15–16 Sept. 1999, bringing flooding rains, high winds, rough seas, and property and ecological damage. The total damage to North Carolina from Hurricane Floyd includes 51 people killed, 7000 homes destroyed, power loss for 500000 people, and more than $1 billion in damage (Kilborn, 1999; National Oceanic and Atmospheric Administration, 1999). Hurricane Floyd is estimated to have destroyed more than 2 million farm animals in North Carolina (Kilborn, 1999). The flooding resulting from Hurricane Floyd caused many hog lagoons to flood or rupture. In addition, severe flooding adversely affected the operation of municipal wastewater treatment plants (WWTPs), releasing untreated wastewater into rivers. Both flooded hog lagoons and flooded WWTPs contributed to nitrogen (N) contamination in the rivers of eastern North Carolina.
Nitric oxide (NO) is a precursor to tropospheric ozone formation and contributes to acid rain. Tropospheric ozone increases the oxidizing capacity of the troposphere, leading to detrimental effects on human health and reduced crop yields, as well as increasing overall maintenance costs for buildings and machinery. Soil is an important contributor of NO to the atmosphere. Nitrogen fertilization of agricultural soils has been shown to increase NO flux from soils (Smith et al., 1997; Tabachow et al., 2001). The effects of prolonged hurricane-induced flooding on NO flux in agricultural areas have not been studied in a well-controlled laboratory environment. This study focuses on hurricane-loaded soils collected from three sites along the bank of the Neuse River in eastern North Carolina and examines the NO flux under controlled laboratory conditions for different levels of water-filled pore space (WFPS). The atmospheric transport and transformation of NO is outside of the scope of this research. Projections of the effect of Hurricane Floyd on select water quality parameters in the Neuse River basin were performed, and the long-term effects of this hurricane-loaded soil on NO flux are discussed.
|
SOIL SAMPLING LOCATIONS
|
|---|
The Neuse River basin extends from the Piedmont region of North Carolina to the mouth of the Pamlico Sound with a drainage area of more than 16000 km2 (6200 mi2) and more than 5148 km (3200 mi) of freshwater streams (North Carolina Department of the Environment and Natural Resources, 2000). Agricultural land comprises 35% of the basin. Nitrogen is a primary pollutant of the Neuse River basin and the North Carolina Department of Environment, Health, and Natural Resources has mandated a 30% reduction of N loading within the basin by 2001. The three sites selected for sampling and NO emissions measurements lie in geographic areas affected differently by Hurricane Floyd (Fig. 1)
representative of the upper, middle, and lower Neuse River basin. Nine soil samples from each site were collected in March and April 2000. Each soil sample was collected approximately 3 to 6 m (10 to 20 ft) from the bank of the Neuse River from soil that is typically not submerged by the Neuse River.
View larger version (94K):
[in this window]
[in a new window]
|
Fig. 1. Three soil sampling locations (rectangles) within the Neuse river basin (shaded) in eastern North Carolina with swine farms represented by dots (adapted from Hogwatch, 2000). Flooding occurred within the river basin east of Interstate Highway 95 to the Atlantic Ocean. Scale: 1" = 60 miles.
|
|
The Wake Forest (Site 1) soil samples were collected at Falls of the Neuse Lake near Wake Forest, NC. This site lies approximately 200 km (125 mi) upstream of the Atlantic Ocean and has an elevation of 91 m (300 ft) above sea level. Wake Forest, NC suffered only minor flooding during Hurricane Floyd, as did most of the North Carolina Piedmont. The Wake Forest soil is a White Store sandy loam (fine, mixed, active, thermic Oxyaquic Vertic Hapludalf), which is a moderately well-drained soil found on 6 to 10% slopes (Cawthorn, 1970). Soil texture was also determined in the laboratory (Table 1) using a hydrometer method. The soil samples were collected from the eastern shore of the Falls of the Neuse Lake reservoir, which was flooded for several days to a height of 0.3 to 0.9 m (1 to 3 ft) above flood level. The majority of septic tanks around the Falls of the Neuse State Park were pumped empty before Floyd.
The Kinston (Site 2) soil samples were collected from Kinston, NC, approximately 56 km (35 mi) upstream of the Atlantic Ocean. The Kinston soil is a Lakeland sand (thermic, coated, Typic Quartzipsamment) that is aggressively drained (Barnhill, 1977). Lakeland sands consist of nearly level to gently sloping soils with high permeability. Flooding from Hurricane Floyd was especially high in this part of North Carolina's coastal plains. At the soil sampling location, the Neuse River has a relatively narrow cross-section, which together with the relative flatness of the surrounding region contributed to significant flooding. During the peak flooding caused by Hurricane Floyd, this stage of the Neuse River exceeded 8.2 m (27 ft) above flood level. The Kinston site is located downstream of the population centers of Raleigh, Smithfield, Clayton, and Goldsboro, and located downstream of a large number of hog farms.
The third sampling location was in New Bern, NC. This site lies at the junction of the Neuse Estuary and the Pamlico Sound. Water was recorded as rising between 0.6 and 1.5 m (2 and 5 ft) above flood level. New Bern sustained only minor flooding due to the holding capacity of Neuse and Trent Rivers and the Pamlico Sound. However, a large portion of the watershed drains through these downstream soils and high concentrations of N contaminants were introduced to the New Bern soils. The New Bern soil is a Dorovan muck (dysic, thermic Typic Haplosaprist) that is a very poorly drained, frequently flooded soil on broad flood plains (Goodwin, 1989).
|
EXPERIMENTAL DESIGN
|
|---|
Nine soil samples were collected from each of the three sites in March and April 2000 for analysis of soil NO flux and for soil characterization (Table 1). The NO flux from each of the nine soil samples at each site was measured in the laboratory at controlled WFPS values of 3, 20, and 40%. Water-filled pore space is an index of the moisture content of soil. Water-filled pore space is useful in indicating the air to water ratio in soil as well as the amount of water available to support microbial activity in soil (Ormeci et al., 1999; Tabachow et al., 2001). Therefore, WFPS is potentially useful in discussing the transfer of NO to the atmosphere via diffusion through water-filled pore spaces, a relatively slow process, or rapidly via advection through air-filled pore spaces. Diffusion is highly variable in rate, depending on the concentration gradient and soil pore structure.
The flooding from Hurricane Floyd began in September 1999 and the hurricane-induced flooding lasted through October 1999. The objectives of this research focus on the long-term effects of hurricane-loaded soil on NO flux. The most recent hurricane to impact the soil at the three sites occurred approximately 6 mo prior to soil collection. Previous research suggests terrestrial recovery from a hurricane can take months to decades (Valiela et al., 1998). Valiela et al. (1998) studied the hurricane effects on terrestrial ecosystems including effects on forests, leaching of salt to soils, and disturbances to terrestrial animals. Additional research indicates that the effect on NO flux from municipal wastewater treatment biosolids–amended soil is often not observed until months after application. Scott et al. (1998) suggest that the application of mineral fertilizer to soil increases NO flux for a period of only 1 to 2 wk, while application of biosolids increases NO flux for a period of several months. Biosolids can provide a source of organic N that is progressively mineralized, prolonging the effect on gaseous N flux. Thus, 6 mo after N loading is a reasonable time to study the soil.
The devastation caused by Hurricane Floyd offers a unique opportunity for the study of NO flux from hurricane-impacted soil. Nitric oxide flux experiments were not conducted on these three soils prior to the hurricane and effects on NO flux cannot be conclusively related to Hurricane Floyd loading. Changes in NO flux are difficult to quantify after individual episodic events due to varying soil and environmental conditions including temperature, WFPS, atmospheric pressure, and wind velocity. This study can provide a baseline for future hurricane-loaded soil research.
The soil samples were collected using an auger at a maximum depth of 30 cm below grade. Each soil sample consisted of approximately 3000 g of soil and was continuously refrigerated at a temperature of 4°C in an airtight bag until needed. Each soil sample was then thoroughly mixed with a stainless steel trowel and sieved to achieve a uniform particle size through a No. 10 U.S. standard sieve (2-mm openings).
Experimental Setup
The dynamic test chamber and experimental setup (Fig. 2)
was developed at Duke University based on standard experimental research procedures employed over the past decade (Peirce et al., 1987; Ormeci et al., 1999; Peirce and Aneja, 2000; Tabachow et al., 2001). The test chamber consists of a glass column enclosed by top and bottom nonreactive Teflon plates incorporating Viton O-rings to maintain an airtight fitting. All gasses enter and exit the test chamber from ports in the Teflon top plate. The soil column was loaded to a height of 11 cm (approximately 1000 g of soil) and compacted for 1 h using a 1-kg weight. After compaction, the 1-kg weight was removed and the Teflon top was inserted onto the test chamber and secured with wing nuts. Teflon tubing connections were made to the zero-grade air, the Teflon stirrer, and the NO analyzer.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2. Experimental setup using dynamic test chamber. Glass column of test chamber is 9.8 cm in diameter by 20.5 cm in height.
|
|
A continuous flow of zero-grade air (National Welders [Charlotte, NC] zero-grade air: 0% hydrocarbons, 0% moisture, 0% NO) was used for all experiments at a constant flow rate of 1.25 L/min based on research performed by Tabachow et al. (2001). A Teflon stirrer was used to thoroughly mix the zero-grade air with gaseous soil emissions within the test cell. The stirrer was operated continuously at an approximate rate of 100 rpm. The test chamber maintained a slight positive pressure throughout the experimental sequence because the influent flow rate of zero-grade air is higher than the effluent flow rate to the NO analyzer. Excess gas within the test chamber was allowed to escape the chamber via the ball-bearing stirrer connector.
The NO analyzer used in this research is a Thermo Environmental Instruments (Franklin, MA) Model 42S Chemiluminescence Low Level NO–NO2–NOx analyzer. The NO analyzer had a range of 0 to 200 parts per billion by volume (ppbv) and was calibrated according to the manufacturer's specifications using a multipoint dynamic calibration. The sample flow rate into the analyzer averaged approximately 1 L/min.
Soil Water-Filled Pore Space Preparation
The WFPS parameter was targeted at three values, 3, 20, and 40%. The 3% WFPS target value was selected because soil that has been air-dried for 24 h results in an approximate WFPS of 3%. The highest WFPS value of 40% was selected based on research performed by Ormeci et al. (1999), which suggests an exponential decline in NO emissions at WFPS > 45%, attributed to limited advective NO gas transport as pore spaces are filled with water and to decreased nitrifying activity. The 20% WFPS value is approximately midway between 3% and 40%. The WFPS was adjusted by either air-drying the soil or wetting the soil. Soil was air-dried at room temperature in either a stainless steel pan or spread on a thick sheet of nonabsorbent paper.
Calculation of Nitric Oxide Flux
The mass balance for NO in the chamber (Kaplan et al., 1988) is given by:
 | [1] |
where [C]o = NO concentration at the inlet of the chamber (ng N/m3); [C]f = NO concentration in the chamber (ng N/m3) measured by the analyzer in ppbv; A' = surface area of the experimental setup, including glass column and Teflon tubing in contact with the headspace atmosphere (m2); A = chamber cross-sectional area (m2); V = volume of the chamber headspace (m3); Q = air flow rate through the chamber (m3/s); J = NO emissions from soil, flux per unit area [ng N/(m2 s)]; L = total loss term encompassing loss of NO on the chamber wall and in tubing, assuming first-order reactions in NO concentration (m/s); and R = chemical production–consumption rate for NO in the chamber [ng N/(m3 s)]. The test chamber and all supporting apparatus are composed of chemically inert materials (e.g., glass and Teflon). Therefore, NO production and NO consumption reactions taking place within the test chamber are not the result of test cell material and gaseous product interactions. Influent into the test chamber is free of NO with the inflow air consisting solely of zero-grade air; therefore, [C]o = 0. The stirrer acts to continuously mix the gasses within the chamber and [C]f is assumed to be equal to the NO concentration everywhere in the chamber headspace.
The NO concentration measured by the analyzer reaches steady state within 20 minutes and the change of NO with time approaches zero thereafter. At steady state, the NO flux is controlled by biological and chemical NO production and by NO transport in the soil. Equation [1] becomes:
 | [2] |
The total loss term, L, represents the loss of NO through reactions with the chamber walls and tubing and with existing reactive oxidants. The influent air into the chamber is zero-grade air and thus existing oxidants are assumed to be negligible. The loss term was determined experimentally by dosing the soil chamber with known amounts of NO when soil was not present in the chamber. L was determined to be negligible (results not shown). Equation [2] reduces to:
 | [3] |
Equation [3] is used to relate the NO concentration measured by the NO analyzer (in ppbv) into a NO flux value [in ng N/(m2 s)]. All experiments were conducted for a minimum of 30 min, with steady-state conditions typically observed within 20 min. Nitric oxide analyzer readings were recorded every minute. Three replicates were conducted for each experimental scenario. The NO flux measurements were observed to be reproducible, with 85% of replicate experiments reproducible within 10% of one another. The 15% of replicate experiments not reproducible within 10% of one another are attributed to natural variations in the soil characteristics.
|
NITRIC OXIDE FLUX RESULTS
|
|---|
Twenty-seven NO flux experiments on soil collected from three locations along the Neuse River basin were conducted (Fig. 3)
. The data on NO flux from the soil ranged from 0.1 to 102.5 ng N/(m2 s). Based on the results of an analysis of variance (ANOVA) statistical test using a 0.05 level of significance, the mean NO flux values at each sample location increased significantly from upstream to downstream sampling locations as follows: Wake Forest [0.5 ng N/(m2 s)], Kinston [17.5 ng N/(m2 s)], and New Bern [77.0 ng N/(m2 s)]. The laboratory results are consistent with unpublished data collected by researchers at North Carolina State University, Raleigh, NC, at the three soil sampling field locations using an on-site dynamic flow-through chamber to measure NO emissions in a mobile laboratory.
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 3. Nitric oxide (NO) flux results for the three sampling locations with the percent water-filled pore space (WFPS) specified. Error bars represent 95% confidence intervals (n = 3).
|
|
|
PROJECTION OF SELECT WATER QUALITY PARAMETERS
|
|---|
Select water quality parameters influenced by major sources of N in floodwaters were projected to develop a better understanding of the impact of flooded soils in the Neuse River basin on NO flux. In eastern North Carolina, two important N sources in floodwaters are municipal wastewater and hog waste (United States Geological Survey, 1999b). Although highly controlled under nonflooded operation, WWTPs and hog lagoons both flooded during Hurricane Floyd (Kilborn, 1999). Extended power outages due to Hurricane Floyd prevented wastewater treatment and millions of liters of untreated and partially treated wastewater were diverted into the Neuse River basin. Due to Hurricane Floyd, hog lagoons were inundated with rainwater and unable to store additional waste. Lagoons filled to capacity with hog waste and rainwater, and floodwaters were unable to store additional hog waste, which was diverted into the Neuse River basin. The feces and urine of other livestock produced in the region, such as poultry and cattle, also impacted the Neuse River basin but to a far less extent than hog waste (Kilborn, 1999). In addition, hog production is the only intensive livestock operation in eastern North Carolina that commonly uses lagoons for treatment and storage of waste.
Determining the flow of water through the Neuse River basin at the three sample locations during the flooding associated with Hurricane Floyd is necessary to estimate the concentrations of N, P, and total solids input into the Neuse River basin by WWTPs and hog farms. The peak flow during Hurricane Floyd at each of the three sample sites was measured by the United States Geological Survey (2000a) before flooding disabled many of the flow gauges. An assumed flow of 5% of this peak was used in the calculations to determine select water quality concentrations. Five percent of the peak flow was chosen as an estimate for the Neuse River basin for the 30 d following Hurricane Floyd.
The purpose of estimating select water quality parameters (N, P, and total solids) is to provide an approximation of the impact of hog farms and WWTPs on the Neuse River basin. This paper focuses on the NO flux from soils, not on precisely estimating the hurricane effects on water parameters. Therefore, a range of values was projected for select water quality parameters. The low-end projection is based on a scenario in which the floodwater impact lasts 5 d. This scenario is based on the observed duration of the flooding in some areas of eastern North Carolina and the anticipated duration of N-impacted floodwater influence. The high-end projection is based on a scenario in which the floodwater impact lasts 30 d. Flood recession, power restoration, and electrical repairs were observed to require 30 d in much of eastern North Carolina after Hurricane Floyd. According to a North Carolina Department of Water Quality report of 24 Sept. 1999, at least a dozen municipal WWTPs in the Neuse River basin were categorized as flooded by Hurricane Floyd 9 d after Hurricane Floyd, including Kinston and Wake Forest (North Carolina Department of the Environment and Natural Resources, 1999b). An updated list generated on 12 Oct. 1999, 28 d after Hurricane Floyd struck North Carolina, indicated that multiple WWTPs in the Neuse River basin were still not fully operational (North Carolina Department of the Environment and Natural Resources, 1999c). These two scenarios vary only in the duration of flooding; therefore, the projected mass of pollutants entering the Neuse River basin will vary within a range. The concentration of the pollutants in both scenarios will not vary as the ratio of mass of contaminant to volume of river water remains constant regardless of the assumed duration of flooding.
Only a few published studies have investigated the effects of hurricanes on coastal watersheds (e.g., Valiela et al., 1998; Mallin et al., 1999) and only a few studies have investigated the effects of swine waste lagoon releases into receiving waters (e.g., Burkholder et al., 1997; Mallin et al., 1997). This study presents the first known published study of hurricane effects on soil NO flux while projecting the impacts of untreated hog waste and municipal wastewater on a coastal watershed.
Projection of the Impact of Municipal Wastewater
Municipalities in each county of the Neuse River basin were identified to project the impact of N, P, and total solids due to WWTP flooding. Municipal population estimates were obtained from the North Carolina Office of State Planning for July 1999 to identify population centers with more than 2500 inhabitants (North Carolina Office of State Planning, 1999), as these cities are typically serviced by municipal WWTPs. Cities with populations under 2500 people were omitted from our analysis as a conservative assumption that these cities contribute relatively small amounts of N to floodwaters. For the purposes of this projection, WWTPs located east of the Interstate Highway 95 were assumed to be non-operational for a limited time after Hurricane Floyd due to flooding and power outages.
The impact of wastewater on the Neuse River basin was projected based on the concentration of wastewater constituents prior to treatment (Table 2). A wastewater outflow of 378 liters (100 gallons) per person per day was assumed (Viessman and Hammer, 1993). A total of 12 cities with a combined population of 175000 people were incorporated into our projection (Table 3). Based on our two flooding scenarios, wastewater production is projected to have ranged from 5.7 to 34.4 billion liters (1.5 billion to 9.1 billion gallons) of wastewater produced and diverted to the Neuse River basin over the course of flooding associated with Hurricane Floyd. Approximately 14000 to 82000 kg (31000 to 180000 lb) of N are projected to have entered the river basin, correlating to the addition of 0.4 mg/L of N. Approximately 3200 to 19000 kg (7000 to 42000 lb) of P, correlating to 0.1 mg/L of P, and 320000 to 1900000 kg (700000 to 4200000 lb) of total solids, correlating to 9.2 mg/L of total solids, are projected to have entered the Neuse River basin due to Hurricane Floyd.
View this table:
[in this window]
[in a new window]
|
Table 3. Projection of select municipal wastewater and swine waste parameters on the Neuse River basin due to Hurricane Floyd. Scenario 1 (low projection) is based on a floodwater impact period of 5 d. Scenario 2 (high projection) is based on a floodwater impact period of 30 d.
|
|
Projection of the Impact of Hog Farms
Large-scale hog production facilities are commonly found in eastern North Carolina, where thousands of hogs are raised in long, shed-like structures. Urine and feces are rinsed into nearby open lagoons to undergo sludge settling and anaerobic breakdown. Each hog lagoon has a finite storage capacity and is susceptible to inundation, breaching, or overtopping (Mallin et al., 1997). The 1997 U.S. Agricultural Census identified 682 hog farms in the eastern North Carolina counties within the Neuse River basin. To project the N loading due to flooding of hog lagoons, the number of hogs in each flooded county and the hog waste generated per animal per day was estimated. The data for hog population and hog waste production were obtained from the 1997 U.S. Census of Agriculture, the 2000 North Carolina Agricultural Chemicals Manual (Barker et al., 1999), and Hogwatch (2000). Hog population data are collected at the county level by the state of North Carolina agricultural statistics program (USDA, 1999) and compiled in the U.S. Census of Agriculture (USDA, 2000). The N, P, and total solids concentrations in hog waste are presented in Table 2. For the purposes of this projection, we assumed that flooding due to Hurricane Floyd filled to capacity all hog lagoons in the Neuse River basin. All hogs were assumed to produce their typical quantity and quality of urine and feces each day, based on mean waste generated, and this hog waste was assumed to enter the Neuse River basin without any treatment. An impact period of 5 to 30 d was assumed, as in the WWTP impact projection. This projection does not consider hog waste from breached lagoons that were reported to occur (Kilborn, 1999; North Carolina Department of the Environment and Natural Resources, 1999a), and therefore may underreport the amount of hog waste entering the basin. In addition, this analysis does not consider runoff from hog waste sprayfields.
According to the North Carolina Department of Agriculture, the swine population in North Carolina was 9.9 million when Hurricane Floyd struck (Mallin, 2000). Of these, more than 2 million hogs were estimated to live in the Neuse River basin. Based on our two flooding scenarios, feces and urine production from these hogs are projected to have ranged between approximately 54000000 and 320000000 liters (14250000 and 85500000 gallons) during the flooding associated with Hurricane Floyd. The quantity of N added to the Neuse River basin from this hog waste is projected to have been between 330000 and 1900000 kg (720000 and 4300000 lb), which correlates to 9.4 mg/L of N entering the basin. Phosphorus additions are projected to have ranged from 110000 to 640000 kg (240000 to 1400000 lb), correlating to 3.1 mg/L of P. Total solids additions are projected to have ranged from 6600000 to 40000000 kg (14500000 to 88000000 lb), or 190 mg/L of total solids.
|
DISCUSSION
|
|---|
Based on ANOVA at a 0.05 level of significance, NO flux was observed to increase significantly from upstream to downstream locations at the three sample sites with the highest NO flux at the most downstream location. The Wake Forest soil was collected the greatest distance from the Pamlico Sound (approximately 200 km [125 mi]). The soil from Kinston was collected at a distance of approximately 56 km (35 mi) from the Pamlico Sound and the soil from New Bern was collected only 1.6 km (1 mi) from the Pamlico Sound.
Previous research has suggested that the concentration and chemical species of N in soil are the primary factors influencing the NO flux from soil (Smith et al., 1997; Jousset et al., 2001; Tabachow et al., 2001). The concentration of nitrite N and nitrate N at all three sites was found to be below detection levels (Table 1). The amount of ammonia, measured as ammonia N in mg/kg on a dry weight basis (mg/kg dry wt.), was found to increase from Wake Forest to Kinston to New Bern (upstream to downstream) from <55.7 to 56.1 to 213, respectively. This increase in ammonia concentration is suggested to be at least partially due to the influence of Hurricane Floyd flooding. Valiela et al. (1998) suggested that the leaching of salt to inland soils following a hurricane releases previously adsorbed ammonium, which transports N to receiving waters and flooded soils. The inundation of soil with floodwaters may have also created reduced conditions in the soil, which may have added to the ammonia concentration detected in flooded soils. The most downstream site, New Bern, was found to have a total organic carbon concentration of 20000 mg/kg dry wt. and total Kjeldahl N concentration of 2360 mg/kg dry wt. These concentrations are an order of magnitude higher than the Wake Forest or Kinston soils. Soil samples were not collected prior to Hurricane Floyd; therefore, it is not possible to analyze quantitatively the hurricane effects on the soil NO flux at these three sites. Water-filled pore space was not observed to influence significantly the NO emissions from these three soils.
Numerous researchers have studied the influence of soil type on NO flux (Peirce and Aneja, 2000; Tabachow et al., 2001). Nitrifying and denitrifying bacteria have been suggested to contribute to NO production in soil (Jousset et al., 2001). Jousset et al. (2001) found that autotrophic nitrification is the major source of NO production in soils compared with respiratory denitrification, chemodentrification, and heterotrophic nitrification. Autotrophic nitrification is an aerobic process that is less likely to occur in the anaerobic muck New Bern soil. Thus, based on soil type, the New Bern soil would be expected to produce the lowest NO flux.
Soil and water quality in eastern North Carolina commonly is subjected to effects from hurricanes. The management of waste generated by livestock and humans can magnify significantly effects on water quality and flooded soil. Only recently have researchers studied animal waste lagoons as a significant source of pollution following a hurricane (Mallin et al., 1999), in part due to the recent increase in large-scale animal waste operations. Legislative efforts in North Carolina over the past several years have been made to ban future hog lagoons from being situated in the 100-yr floodplain and reduce spraying of lagoon liquid onto rain-saturated floodplains (Mallin, 2000).
|
CONCLUSIONS
|
|---|
(i) The NO flux measured from soil samples collected from the downstream regions of the Neuse River basin was observed to be significantly higher, based on ANOVA at a 0.05 level of significance, than from upstream soils even months after the flooding occurred. The mean (95% confidence) NO flux from the Wake Forest, Kinston, and New Bern soils was 0.5, 17.5, and 77.0 ng N/(m2 s), respectively. These findings suggest that the increased N loading from floodwaters impacted with municipal wastewater and hog waste enhanced the NO emissions from the downstream areas of the Neuse River basin. No data was available to distinguish the incremental effects associated with flooding due to Hurricane Floyd.
(ii) Hurricane Floyd impacted the Neuse River basin by inducing severe flooding, which is projected to have adversely increased water quality parameters. Wastewater treatment plants serve approximately 175000 people in the Neuse River basin and flooding and power outages are projected to have added approximately 5.7 to 34.4 billion liters (1.5 billion to 9.1 billion gallons) of untreated wastewater into the basin. Approximately 14000 to 82000 kg (31000 to 180000 lb) of N (0.4 mg/L of N), 3200 to 19000 kg (7000 to 42000 lb) of P (0.1 mg/L of P), and 320000 to 1900000 kg (700000 to 4200000 lb) of total solids (9.2 mg/L of total solids) are projected to have entered the basin from WWTP flooding. Hog lagoons for approximately 2 million swine were inundated with rain and floodwaters, allowing a projected 54000000 to 320000000 liters (14250000 to 85500000 gallons) of untreated hog waste to enter the Neuse River basin. Approximately 330000 to 1900000 kg (720000 to 4300000 lb) of N (9.4 mg/L of N), 110000 to 640000 kg (240000 to 1400000 lb) of P (3.1 mg/L of P), and 6600000 to 40000000 kg (14500000 to 88000000 lb) of total solids (190 mg/L of total solids) are estimated to have entered the Neuse River basin due to hog waste lagoon flooding. The resulting increased concentrations of these water quality parameters are suggested to have affected the NO flux from flooded soils along the Neuse River basin.
(iii) Topics of future research suggested by this research are: (a) continued study of the effects of soil characteristics and soil contaminants on NO flux; (b) impact of NO on air quality; (c) the agricultural value of N lost to receiving water bodies and to the airshed as NO; (d) investigation of the influence of varying types of soil on ammonia adsorption and NO emissions; and (e) improved management of the disposal of animal waste stored in lagoons in hurricane- and flood-prone areas.
|
ACKNOWLEDGMENTS
|
|---|
The Culpeper Foundation, in particular Professor Marie Lynn Miranda of the Nicholas School of the Environment, and the National Science Foundation, in particular Dr. Edward Bryan from the Division of Bioengineering and Environmental Systems, made this work possible through their financial support. The writers also wish to acknowledge Dr. Banu Ormeci for her assistance in reviewing the manuscript, and Dr. Viney Aneja and Paul Roelle of the Department of Earth, Marine, and Atmospheric Science at North Carolina State University, Raleigh, NC for their assistance in collecting field NO data.
|
REFERENCES
|
|---|
- Barker, J.C., S.C. Hodges, and C.R. Campbell. 1999. Livestock manure production rates and nutrient content. 2000 North Carolina agricultural chemicals manual. The College of Agric. and Life Sciences, North Carolina State Univ., Raleigh.
- Barnhill, W.L. 1977. Soil survey of Lenoir County, North Carolina. USDA Soil Conserv. Serv., Washington, DC.
- Brady, N.C., and R.R. Weil. 1999. The nature and properties of soils. 12th ed. Prentice Hall, Upper Saddle River, NJ.
- Burkholder, J.M., M.A. Mallin, H.B. Glasgow, L.M. Larsen, M.R. McIver, G.C. Shank, N. Deamer-Melia, D.S. Briley, J. Springer, and B.W. Touchette. 1997. Impacts to a coastal river and estuary from rupture of a large swine waste holding lagoon. J. Environ. Qual. 26:1451–1466.[Abstract/Free Full Text]
- Cawthorn, J.W. 1970. Soil survey of Wake County, North Carolina. USDA Soil Conserv. Serv., Washington, DC.
- Goodwin, R.A. 1989. Soil survey of Craven county, North Carolina. USDA Soil Conserv. Serv., Washington, DC.
- Hogwatch. 2000. Environmental Defense's central source for information and action on industrial hog farming in North Carolina. Available online at www.hogwatch.org (verified 16 July 2001). Environmental Defense, New York.
- Jousset, S., R.M. Tabachow, and J.J. Peirce. 2001. Soil nitric oxide emissions from nitrification and denitrification. J. Environ. Eng. (Reston, VA) 127:322–328.
- Kaplan, W.A., S.C. Wofsy, M. Keller, and J.M. Da Costa. 1988. Emission of NO and deposition of O3 in a tropical forest system. J. Geophys. Res. 93(D2):1389–1395.
- Kilborn, P.T. 1999. Hurricane reveals flaws in farm law. The New York Times. 17 Oct.
- Mallin, M.A. 2000. Impacts of industrial animal production on rivers and estuaries. Am. Scientist 88:26–37.
- Mallin, M.A., J.M. Burkholder, M.R. McIver, G.C. Shank, H.B. Glasgow Jr., B.W. Touchette, and J. Springer. 1997. Comparative effects of poultry and swine waste lagoon spills on the quality of receiving streamwaters. J. Environ. Qual. 26:1622–1631.[Abstract/Free Full Text]
- Mallin, M.A., M.H. Posey, G.C. Stiank, M.R. McIver, S.H. Ensign, and T.D. Alphin. 1999. Hurricane effects on water quality and benthos in the Cape Fear watershed: Natural and anthropogenic impacts. Ecol. Applic. 9:350–362.
- National Oceanic and Atmospheric Administration. 1999. Climate-watch, September 1999. Available online at www.ncdc.noaa.gov/ol/climate/extremes/1999/september/extremes0999.html (verified 16 July 2001). NOAA, Washington, DC.
- North Carolina Department of the Environment and Natural Resources. 1999a. Updated summary of reported environmental problems associated with Hurricane Floyd. 17 Sept. 1999. Available online at www.enr.state.nc.us/newsrels/floyd2.htm (verified 16 July 2001). North Carolina Dep. of the Environ. and Natural Resour., Div. of Water Resour., Raleigh.
- North Carolina Department of the Environment and Natural Resources. 1999b. NPDES municipal WWTP's that have been flooded by Hurricane Floyd. 24 Sept. 1999. Available online at www.enr.state.nc.us/floyd/dams3.htm (verified 16 July 2001). North Carolina Dep. of the Environ. and Natural Resour., Div. of Water Resour., Raleigh.
- North Carolina Department of the Environment and Natural Resources. 1999c. NPDES municipal WWTP's that have been flooded by Hurricane Floyd. 12 Oct. 1999. Available online at www.enr.state.nc.us/floyd/wwtp2.htm (verified 16 July 2001). North Carolina Dep. of the Environ. and Natural Resour., Div. of Water Resour., Raleigh.
- North Carolina Department of the Environment and Natural Resources. 2000. Neuse River basin. Available online at www.dwr.ehnr.state.nc.us/~rivers/neuse.htm and ftp://h2o.enr.state.nc.us/pub/NonDischarge/ (verified 16 July 2001). North Carolina Dep. of the Environ. and Natural Resour., Div. of Water Resour., Raleigh.
- North Carolina Office of State Planning. 1999. State demographics. Available online at www.ospl.state.nc.us/demog (verified 16 July 2001). North Carolina Office of State Planning, Raleigh.
- Ormeci, B., S.L. Sanin, and J.J. Peirce. 1999. Laboratory study of NO flux from agricultural soil: Effects of soil moisture, pH, and temperature. J. Geophys. Res. Atmospheres 104(D1):1621–1629.
- Peirce, J.J., and V.P. Aneja. 2000. Nitric oxide emissions from engineered soil systems. J. Environ. Eng. (Reston, VA) 126:225–232.
- Peirce, J.J., G. Sallfors, and K. Ford. 1987. Differential flow patterns through compacted clay. Geotech. Testing J. 10:218–222.
- Scott, A., B.C. Ball, I.J. Crichton, and M.N. Aitken. 1998. Nitrous oxide emissions from grassland amended with sewage sludge. Soil Use Manage. 14:55.
- Smith, K.A., I.P. McTaggart, and H. Tsuruta. 1997. Emissions of N2O and NO associated with nitrogen fertilization in intensive agriculture, and the potential for mitigation. Soil Use Manage. 13:296–304.
- Tabachow, R.M., J.J. Peirce, and S. Jousset. 2001. Nitric oxide emissions from fertilized and biosolids amended soil. J. Environ. Eng. (Reston, VA) 127:517–523.
- USDA. 1999. 1997 Census of Agriculture. Available online at www.nass.usda.gov/census (verified 16 July 2001). USDA Natl. Agric. Statistics Serv., Washington, DC.
- USDA. 2000. North Carolina agricultural census. Available online at http://govinfo.library.orst.edu/cgi-bin/ag-state?North+Carolina (verified 16 July 2001). USDA, Washington, DC.
- United States Geological Survey. 1999a. Hurricane Floyd beats out Fran with record flood levels on North Carolina rivers. 21 Sept. 1999. USGS, Raleigh, NC.
- United States Geological Survey. 1999b. USGS scientists tracking environmental damage from Floyd...Heavy flooding caused heavy pollution. 23 Sept. 1999. USGS, Reston, VA.
- Valiela, I., P. Peckol, C. D'Avanzon, J. Kremer, D. Hersh, K. Foreman, K. Lajtha, W.R. Geyer, T. Isaji, and R. Crawford. 1998. Ecological effects of major storms on coastal watersheds and coastal waters: Hurricane Bob on Cape Cod. J. Coastal Res. 14:218–238.
- Viessman, W., and M.J. Hammer. 1993. Water supply and pollution control. HarperCollins, New York.
TOP
ABSTRACT
INTRODUCTION
