| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Animal Sciences Dep., Purdue Univ., West Lafayette, IN 47907
* Corresponding author (heber{at}purdue.edu)
Received for publication November 20, 2001.
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: AU, animal unit • BCFC, buoyant convective flux chamber • BIW, equivalent concentration of n-butanol in water • DT, dilutions to threshold • HT, hedonic tone • OC, odor concentration • OCE, odor concentration normalized to European odor units • ODCb, odor detection concentration of n-butanol gas • OU, odor unit • OUE, European odor unit, equivalent to 123 µg of n-butanol • VSLR, volatile solids loading rate
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Measurements of the character and quantity of odor releases are also needed to evaluate odor effects through science-based setback models (Lim et al., 2000). Odor is evaluated by determining odor concentration (OC), intensity, and hedonic tone (HT) (Lim et al., 2001). An odor unit (OU) is defined as the amount of odorant(s) in 1.0 m3 of odorous gas at the panel odor detection concentration (European Committee for Standardization, 2002) and OC is defined in terms of OU m-3. An odor emission rate (OU s-1) results when OC is multiplied by volumetric airflow rate in m3 s-1 (Heber et al., 2002).
Anaerobic treatment systems (lagoons) are used widely for practical treatment and storage of swine manure (National Engineering Handbook, 1999; Westerman et al., 1990). Lagoons are typically earthen basins, and rely on bacteria to stabilize organic material (Pork Industry Handbook, 1998). Lagoons are relatively simple to operate and maintain, and are relatively inexpensive compared with other treatment methods (American Society of Agricultural Engineers, 1997). Lagoons become more odorous when overloaded due to sludge buildups, infrequent additions of large amounts of manure, and cold weather (Ritter, 1989).
The traditional wisdom of agricultural engineers has been that a properly designed and operated anaerobic lagoon will have minimal odor problems, except in spring when lagoon temperature rises and bacterial action increases (Hamilton and Cumba, 2000). However, all lagoons generate some odor; the quantity has not been well documented (Heber et al., 2002). Lagoon odor can be reduced by maintaining adequate dilution and improving loading uniformity by introducing smaller amounts of manure more frequently. Anaerobic lagoons are designed on the basis of volatile solids loading rate (VSLR) (National Engineering Handbook, 1999). The acceptable VSLR varies from one location to another since the rate of solids decomposition in anaerobic lagoons is a function of temperature.
The design VSLR for anaerobic lagoons located in southern Indiana and Illinois is 76.1 g d-1 m-3, 64.1 g d-1 m-3, and 72.1 g d-1 m-3 according to the Midwest Plan Service (2000), American Society of Agricultural Engineers (1997), and the National Engineering Handbook (1999), respectively. Sensory-based determinations of lagoon odor release as affected by VSLR would be helpful in guiding design and management of lagoons, but such measurements are lacking.
Several methods of measuring odor and gas release from surfaces have been described in previous literature (Lindvall et al., 1974; Smith and Watts, 1994; Jiang et al., 1995; Schmidt et al., 1999). A convective flux chamber (CFC) is an open-bottom enclosure placed over emitting surfaces, with ambient or filtered air blown or drawn through it to mix and transport gases away from the emitting surface. Concentrations of both incoming and outgoing air streams should be measured when ambient air is used (Smith and Watts, 1994). Release of NH3 was calculated as the product of the difference between inlet and outlet air concentrations and the volume of air passing through a modified Lindvall CFC (Misselbrook et al., 1998). A CFC was also used to evaluate odor and H2S release from manure storages and feedlots (Schmidt et al., 1999). Concentrations in the CFC exhaust air were measured and releases were calculated based on simulated bulk wind speeds that ranged from 0.19 to 1.14 m s-1. Using the same CFC, odor samples were collected at 19 animal manure storage sites by Jacobson et al. (1999) during spring, summer, and fall. Hydrogen sulfide (H2S) and OCs were measured with an electrochemical sensor and a dynamic olfactometer, respectively.
A BCFC was used to measure seasonal release of atmospheric ammonia nitrogen (NH3–N) from an anaerobic swine lagoon during different seasons (Aneja et al., 2000). The BCFC was placed at randomly selected locations and compressed zero-grade air was pumped through the BCFC at constant flow.
Hobbs et al. (1998) designed and constructed a laboratory CFC to measure odor and gas release from 200 L of stirred slurry samples under controlled conditions and a wind speed of 4 m s-1. Gaseous compounds were measured with gas chromatography–mass spectrometry, and odor concentration was evaluated with a dynamic olfactometer. The most dominant odor compounds observed belonged to the sulfide, volatile fatty acid, phenolic, and indolic chemical groups. The mean odor release from the stirred slurry samples (surface area = 1.0 m2, wind speed = 4.0 m s-1) was 1.35 x 106 OU min-1 (2.25 x 104 OU s-1 m-2).
A new BCFC was designed, constructed, and tested at Purdue University (Heber et al., 2002). Repeatable odor and gas release measurements with the BCFC were demonstrated based on laboratory and field tests. Odor release from a 2.4-m-deep primary anaerobic lagoon, with the top layer surface-aerated, was measured at a 6000-head swine finishing facility (Heber et al., 2002). Odor release measured at the lagoon was 1.7 OU s-1 m-2 with a simulated wind speed of 1.1 m s-1.
Some methods for assessing odor emitted from agricultural and municipal wastewaters rely on H2S concentration to surrogatively represent odor strength, because H2S is typically an important component of malodors. Also, H2S measurements are easier, more reproducible, and less expensive than olfactometry. Stuetz et al. (1999) concluded that, although H2S correlates better to odor than other compounds, the correlation is weak.
The objectives of this research project were to (i) determine releases of odor, NH3, H2S, CO2, SO2, and NO from the first stage of anaerobic treatment systems for swine, and (ii) evaluate effects of lagoon loading rates and slurry characteristics on releases of odor, NH3, H2S, CO2, SO2, and NO.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Buoyant Convective Flux Chamber
The BCFC (Heber et al., 2002) covered 0.74 m2 of lagoon surface over which air was blown at approximately 1 m s-1. The BCFC was surrounded by rigid waterproof insulation to cause enough buoyancy to keep the top 0.17 m of the BCFC floating above water. The inside walls and ceiling were lined with stainless steel (Fig. 1)
. Air followed a 0.31-m-wide, horizontal, hairpin path and the total length of the airflow path across the exposed liquid surface was 2.4 m.
|
|
Air samples at the BCFC inlet and outlet were simultaneously drawn into chemically inert 50-L polyvinyl fluoride (Tedlar) bags through Teflon tubing. A small diaphragm pump (AirPro Model 6000D; BIOS International, Butler, NJ) evacuated air from a 114-L rigid drum causing an initially collapsed bag inside the drum to inflate in about 10 min at an airflow rate of 5 L min-1. Negative pressure in the drum caused air to enter the sampling bag directly without flowing through the pump. Inside surfaces of the BCFC were cleaned with alcohol between visits. Air sampling tubes were flushed between visits with compressed air or nitrogen (N2) to purge the tubes of residual odor.
Air entering the BCFC was sampled because the gas absorption device could not remove 100% of the ambient odor on a single pass, and there may have been some odors released from the surfaces of air supply ducts. To minimize contamination of the air transport system, the air supply unit was placed upwind of the lagoon and as far away as possible from exhaust fans of nearby swine buildings. Two downwind air samples were simultaneously collected at the downwind berm, about 1.0 m laterally from the edge of the lagoon, and at a height of 1.0 m above the top of the berm. Air samples at the berm provided a general indication of ambient air quality near the lagoon and were compared with measured background concentrations (BCFC inlet samples).
Sampling bags were either new or reused once. New bags, preflushed once with compressed air or N2, were used for sampling visits A1, A2, A5, B1, B4, and B5. These bags were reused for subsequent visits after being filled and flushed at least three times. The BCFC was operated at least 10 min before collecting odor samples. Each bag was preconditioned by filling it to about one-quarter capacity with sample air and emptying it before collecting the actual sample. The sampled bags were immediately inserted into 76-µm-thick black plastic bags, and placed into an opaque plastic tub to minimize exposure to sunlight and sudden temperature changes. Enough space between the bags was always allowed during transport to prevent mechanical damage.
Effluent samples were collected from four or five locations along the lagoon edges. Samples were obtained by lowering a sampling probe (a 600-mL plastic bucket attached to a 1.9-cm-diameter, 2-m-long PVC pipe) into the lagoon from the shore. Influent samples were collected either from the buildings or directly at the pipe inlets to the lagoons during flushing. Samples of each type were poured into a bucket and thoroughly mixed. A subsample of each mix was collected and stored in a sealed 237-mL plastic bottle, placed in a Styrofoam container with ice, and transported to the laboratory.
Air temperature was measured by locating a thermocouple at the BCFC outlet air sampling point. Lagoon temperature was measured at about 5 cm below the water surface by attaching a thermocouple to the lower edge of the BCFC. Both temperature readings were recorded during collection of each sample.
Odor Evaluation
The odor dilution to threshold (DT) of an air sample is the dilution factor required to reduce its concentration to that which cannot be distinguished from odorless air by 50% of an odor panel (Heber et al., 2002). Odor DTs were measured with a dynamic dilution forced-choice olfactometer (AC'SCENT International Olfactometer; St. Croix Sensory, Lake Elmo, MN) and an odor panel consisting of eight trained human subjects (American Society for Testing and Materials, 1981; Lim et al., 2001). The olfactometer delivered a precise mixture of sample and dilution air to the active subject through a Teflon-coated presentation mask at a flow rate of 20 L min-1. The dilution factor of the mixture was defined as the ratio of total diluted sample flow volume to the odor sample flow volume. Olfactometer airflow rates were calibrated before and after each odor evaluation session with a precision airflow calibration device (GILIBRATOR-2; Sensidyne, Clearwater, FL).
The olfactometer presented an ascending series of concentrations (step factor = 2) to each subject starting with an extremely high dilution factor. The subject sniffed three sequential sample coded gas streams at each dilution factor, with only one gas stream randomly assigned to have the odor. The subject selected the presentation that was "different" and suspected to contain the odor (American Society for Testing and Materials, 1991), and declared whether the selection was a "guess" (no perceived difference), "detection" (selection was different from the other two), or "recognition" (selection smelled like something). Lower and lower sample dilutions (50% reductions) were presented to each subject until the sample was correctly detected and recognized.
An individual best-estimate DT was calculated by taking the geometric mean of the last nondetectable dilution factor and the first detectable dilution factor. Retrospective screening of extraneous individual DTs was applied to the panel DT, which was calculated as the geometric mean of individual DTs. Odor concentration (OU m-3) was numerically equivalent to the panel DT. All averages of odor concentrations and releases were reported as geometric means because they typically exhibit lognormal distributions (European Committee for Standardization, 2002).
To assess subject performance, a reference odorant (40–58 µL L-1 n-butanol in nitrogen) was evaluated identical to other samples during each odor session. In accordance with the EN 13725 standard (European Committee for Standardization, 2002), the evaluations were used to assess subject performance by calculating odor detection concentration (ODC) of the n-butanol, which is the concentration at the detection threshold. The n-butanol ODC of each panel was therefore calculated with Eq. [1]:
 | [1] |
The European standard requires the mean ODCb of the last 12 samples to be between 20 and 80 nL L-1, and a log standard deviation that is smaller than 2.3, for each subject. Most European olfactometry laboratories follow this n-butanol performance criterion to achieve more accurate and repeatable measurements (Sneath and Clarkson, 2000). However, most U.S. laboratories have not typically used reference odorants since U.S. standards do not require them.
Since an n-butanol sample of known concentration was analyzed for each odor session in this study, a corrected odor concentration was determined based on panel sensitivity. Given that one European odor unit (OUE) = 123 µg n-butanol, and 1.0 OUE m-3 = 40 nL L-1 (European Committee for Standardization, 2002), a normalized OC (OCE) was calculated with Eq. [2]:
 | [2] |
Odor intensity is the relative perceived psychological strength of an odor at a suprathreshold concentration (McGinley and McGinley, 2000). Odor intensity grows as a power function of the stimulus odor (Stevens, 1957) and follows Eq. [3]:
 | [3] |
Standardized n-butanol solutions were used to generate a static odor reference scale (American Society for Testing and Materials, 1999) (Table 2). The static reference scale consisted of five concentrations of n-butanol in water with a geometric interval (3x series progression) between each value. A small glass funnel was used to present the sampled air to a subject while the bag was compressed. Each subject judged the intensity of a sample by objectively matching it to the intensities of the known n-butanol concentrations (American Society for Testing and Materials, 1999). The results were reported as equivalent concentrations of n-butanol in water (µL L-1 BIW).
|
Measurement of Gas Concentrations
Concentrations of CO2, H2S, SO2, NH3, and NO were measured in each air sample with gas analyzers described in Table 3 and used in previous studies (Heber et al., 2001). Each analyzer was calibrated twice weekly with certified calibration gases.
|
Calculation of Odor and Gas Releases
The release of a gaseous pollutant was determined by multiplying BCFC airflow rate by the concentration difference between the BCFC inlet and outlet. The BCFC airflow rate was determined by the cross-sectional area of the hairpin path above the water surface and the constant surface air velocity of 1 m s-1 as measured with a hot-wire anemometer (Heber et al., 2002). The release, E, was the transfer of odor or gases from the liquid surface into the atmosphere. It was calculated by dividing the BCFC emission rate by the area covered, AS (0.74 m2):
 | [4] |
The average odor release of a sampling visit was the geometric mean of four individual release measurements. The individual odor release was arbitrarily set equal to 0.1 OU s-1 m-2 when a negative or zero BCFC release was calculated because of an inlet concentration that was greater than or equal to the outlet concentration. The 0.1 value was used to facilitate the calculation of geometric mean values, since zero or negative values cannot be included in geometric mean calculations. Negative or zero releases of major gases were not adjusted.
Analysis of Lagoon Influent and Effluent Samples
Using standard methods, lagoon influent and effluent samples were analyzed for pH, total Kjeldahl nitrogen, ammoniacal nitrogen (NH4–N), total solids, and volatile solids. Total Kjeldahl N was determined by the micro-Kjeldahl nitrogen method of Nelson and Sommers (1972), and NH4–N was determined with the steam distillation method of Bremner and Keeney (1965). The total solids content was analyzed gravimetrically at 90°C. Volatile fatty acid determinations were conducted with gas chromatographic methods described by Playne (1985). In this procedure 8 g of manure sample was mixed with 2 mL of 25% metaphosphoric acid and incubated at a room temperature for 30 min. The samples were centrifuged at 12 100 x g at 4°C for 10 min in a centrifuge with a JA-20 rotor (Model J-21C; Beckman Coulter, Fullerton, CA). The supernatants were drawn off by pipette and frozen for a minimum of 24 h before being passed through a 25-mm, 0.2-µm membrane filter (Supor-200; Pall Corporation, Port Washington, NY). One microliter of the filtered samples was injected into the gas chromatograph (Model 3700; Varian, Palo Alto, CA) for determination of volatile fatty acids with an oven temperature of 155°C and injector and detector temperatures of 200°C each.
Trace Gas Sampling and Analysis
Volatile organic compounds in the sampled air were measured with gas adsorption traps that were constructed from 2.2-mm-i.d., silica-lined, deactivated stainless steel (SilcoSteel Type 304; Restek Corporation, Bellefonte, PA). The traps were sonicated and rinsed in acetone, and packed with Tenax TA 60/80, an adsorbent polymer resin (Alltech Associates, Deerfield, IL). Glass wool placed into both ends of the trap kept the resin from dislodging from the trap. Each end was then fitted with a stainless steel cap to prevent adsorption of compounds from ambient air.
To ensure that the Tenax resin was free from residue, all traps were baked for 60 to 120 min at 220°C in a batch of seven or eight traps at a time with about 20 mL min-1 of N2 flowing through each trap. One or more traps from each batch were randomly selected and analyzed to ensure that the resin was residue-free.
Sample air was drawn through the traps with a vacuum pump (SKC, Eighty Four, PA), which sampled four bags simultaneously at a mean flow rate of 6.9 mL min-1 per trap. Trap airflows were measured with a precision mass flow meter (Digital Flow Check-HR; Alltech Associates) at the beginning and end of each sampling run. To minimize subsequent adsorption of ambient odors into the traps, they were sealed with stainless steel caps and stored in Ziploc bags (SC Johnson, Racine, WI) at -6°C until the time of analysis.
Gaseous compounds adsorbed in each trap were analyzed with an environmental gas chromatograph (Model 8610C; SRI Instruments, Torrance, CA) equipped with a 30-m x 0.53-mm capillary wax column, and a flame ionization detector to analyze target compounds. The target analytes were acetic acid, propionic acid, n-butyric acid, i-butyric acid, n-valeric acid, i-valeric acid, phenol, p-cresol, indole, and skatole (Zahn et al., 1997). The carrier gas was helium with a flow rate of 18 mL min-1 at 34.5 kPa pressure. Oven temperature started at a constant 40°C for 5 min, followed by three ramp increases in succession; from 40 to 110°C at 10°C min-1, 160°C at 5°C min-1, and 190°C at 10°C min-1. The thermal desorber temperature was 260°C and was reached within 60 s, at which time the compounds were injected into the column. The injector valve, held at a temperature of 180°C, remained open for 6 min. The flame ionization detector temperature was 200°C and the flow rates of hydrogen and air to the flame ionization detector were 20 and 250 mL min-1, respectively. Calibration of the gas chromatograph was conducted by spiking clean traps with known amounts of target analytes (liquid form) in 1.0 µL of solution and desorbing them onto the column using the same procedure.
|
RESULTS AND DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Odor and Gas Characteristics
The overall geometric mean odor concentrations of BCFC inlet and outlet samples were 99 and 155 OU m-3 (168 and 262 OUE m-3), respectively (Table 4), similar in magnitude to the geometric mean of 199 OU m-3 measured in exhaust air of swine nursery buildings (Lim et al., 2001). Considerable variation in OC was observed during each sampling visit (Fig. 3)
. Inherent temporal and spatial variation in odor release and lack of precision of OC determinations probably contributed to this variation. To minimize nuisance potential, some atmospheric dilution would have been required to reduce berm concentrations, which ranged from 16 to 219 OU m-3, to the property line limits of 7 to 15 OU m-3 used by several states in the USA (Jacobson et al., 2001).
|
|
Overall mean NH3 concentrations were 3.0, 4.5, and 3.8 mg m-3, and overall mean H2S concentrations were 171, 256, and 143 µg m-3 for inlet, outlet, and berm samples, respectively. During sampling visits, mean outlet NH3 concentrations ranged from 2.4 to 7.5 mg m-3 (Fig. 4) and mean outlet H2S concentrations ranged from 57 to 555 µg m-3 (Fig. 5) .
|
|
Although mean gas concentrations of the BCFC outlet were higher than the inlet, there were some inlet–outlet sample pairs that had higher inlet concentrations. This resulted in the calculation of a "negative" release. Among 48 paired samples, there were 5, 10, 7, and 9 negative releases calculated for NH3, H2S, CO2, and SO2, respectively. The possible causes were random errors in the sampling–measurement procedure coupled with low actual release and potential absorption of gas into the water.
Correlation coefficients between OC, intensity, and HT, and concentrations of NH3, H2S, CO2, and SO2 for all BCFC inlet and outlet samples are presented in Table 5. Results indicated that OCE was directly proportional to intensity and inversely proportional to HT (P < 0.05), similar to evaluations of swine nursery odor in a previous study (Lim et al., 2001). The correlations also showed that OCE was directly proportional to H2S and CO2 concentrations (P < 0.05) and inversely proportional to SO2 concentration (P < 0.05). Whereas little correlation between OC and H2S was found in previous studies of livestock manure slurries and municipal sewage (Jacobson et al., 1999; Stuetz et al., 1999), this study, which normalized OCs to a reference odorant, indicated a statistically significant relationship between H2S and OC.
|
|
|
|
Whereas H2S release data for lagoons was not found in the literature, the mean H2S release (5.7 µg s-1 m-2) in this study was similar in magnitude to that of undiluted stored manure in buildings, earthen basins, and lab reactors. The mean release was near the lower end of releases measured by Jacobson et al. (1999) and Schmidt et al. (1999) from swine manure basins. It was considerably lower than those of stirred slurries (Hobbs et al., 1998, 1999), which might represent an extreme case, since stirred manure emits large quantities of H2S in a short time (Patni and Clarke, 1991). Release of H2S from swine manure tends to be more variable than other gases (e.g., NH3) because of burst releases (Ni et al., 2000b, c).
The mean CO2 release was an order of magnitude smaller than CO2 released from stirred pig slurry (Hobbs et al., 1999), but similar to releases from underfloor pits in emptied swine finish buildings (Ni et al., 2000a) (Table 4). Carbon dioxide is an important component of biogas, produced by anaerobic digestion. Biogas usually contains 40 to 60% CO2.
The mean SO2 release of 0.5 µg s-1 m-2 was greater that the mean release of 0.1 µg s-1 m-2 from laboratory reactors (Ni et al., 2000c). However, surface air speeds in the reactors were much lower than the 1.0 m s-1 air speed in the BCFC.
Effect of Loading Rate
According to the analysis of variance, H2S concentration, OC, intensity, and HT at the BCFC outlet were greater (P < 0.05) at Lagoon A than B (Table 4). The releases of NH3, H2S, and CO2 were 2.7, 4.0, and 3.3 times greater (P < 0.05) at Lagoon A. Since VSLR of Lagoon A was 2.6 times larger than Lagoon B (Table 1), loading rate was shown to have a significant effect on the releases of these gases. It is not known from the data presented whether the significant differences between Lagoons A and B were entirely attributable to higher VSLR; other possible factors included the type of manure collection system and associated frequency of lagoon loading (Table 1).
The geometric mean OCs in the BCFC outlet and at the berm were 300 and 238% higher (P < 0.05) at Lagoon A than at Lagoon B, respectively. While the mean release values of 2.75 and 1.91 OUE s-1 m-2 suggested a similar trend for odor release (Table 4), the difference was not statistically significant (P > 0.05). However, it is reasonable to assume that a lagoon with higher VSLR will release more odors because of higher solids content and bacterial activities. The lack of significance was probably due to relatively large uncertainty associated with estimated VSLR and measured odor release coupled with small numbers of samples and tested lagoons.
Lagoon emission rates can be calculated if uniform flux or release from the entire surface is assumed. The live mass specific emission rates of NH3, H2S, and CO2 were 525, 33, and 4649 mg s-1 AU-1 at Lagoon A as compared with 537, 23, and 3908 mg s-1 AU-1 at Lagoon B. Whereas the releases of NH3, CO2, and H2S were higher (P < 0.05) at Lagoon A, the live mass specific emission rates were not significantly different (P > 0.05). The implication of this observation is that greater dilution achieved with larger lagoon sizes reduces gas releases, but actually increases the total emission from the lagoon.
Effect of Slurry Characteristics
The mean pH values of lagoon influent (n = 4) and effluent (n = 12) samples were 7.9 and 8.1, respectively (Tables 8 and 9). The effluent pH was similar to values reported by Harper et al. (2000) and Aneja et al. (2000). The total solids of Lagoon A effluent was 33% higher than Lagoon B (P < 0.05), but volatile solids were similar. Odor releases appeared (inconclusively) to be related to total solids, which were 2.7 OUE s-1 m-2 for 0.45%, and 1.9 OUE s-1 m-2 for 0.34% total solids at Lagoons A and B, respectively.
|
|
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in JEQ:
This article has been cited by other articles:
|
|
 |
|
  Y.-C. M. Lo, J. A. Koziel, L. Cai, S. J. Hoff, W. S. Jenks, and H. Xin Simultaneous chemical and sensory characterization of volatile organic compounds and semi-volatile organic compounds emitted from swine manure using solid phase microextraction and multidimensional gas chromatography-mass spectrometry-olfactometry. J. Environ. Qual., March 1, 2008; 37(2): 521 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Conn, E. Topp, and G. Lazarovits Factors Influencing the Concentration of Volatile Fatty Acids, Ammonia, and Other Nutrients in Stored Liquid Pig Manure J. Environ. Qual., January 25, 2007; 36(2): 440 - 447. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H. Loughrin, A. A. Szogi, and M. B. Vanotti Reduction of Malodorous Compounds from a Treated Swine Anaerobic Lagoon J. Environ. Qual., January 3, 2006; 35(1): 194 - 199. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
