Environ. Sci. Technol. 2010, 44, 3442–3448
Endotoxin Levels at Swine Farms Using Different Waste Treatment and Management Technologies GWANGPYO KO, * ,† OTTO D. SIMMONS, III, ‡,¶ C H R I S T I N A A . L I K I R D O P U L O S , ‡,| LYNN WORLEY-DAVIS,§ C. M. WILLIAMS,§ AND MARK D. SOBSEY‡ Department of Environmental Health and Institute of Health and Environment, Department of Biological Sciences and Institute of Microbiology, Seoul National University, Seoul, Korea, Department of Environmental Sciences and Engineering, Gillings School of Global Public Health, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, and Department of Poultry Science and Animal and Poultry Waste Management Center, North Carolina State University, Raleigh, North Carolina 27695
Received August 31, 2009. Revised manuscript received February 11, 2010. Accepted March 10, 2010.
Concentrated animal feeding operations (CAFOs) are a major source of airborne endotoxins, which are air pollutants that can cause adverse health effects to both on-site farmers and neighbors. Release of airborne endotoxins to the environment can be reduced using proper waste treatment and management technologies. In this study, the levels of endotoxins released from two swine CAFOs using conventional lagoon-sprayfield technology were compared to those of 15 farms using various alternative waste management technologies in North Carolina. Over a 2-year period, 236 endotoxin samples were collected from the 17 farm units and analyzed using the Limulus amebocyte lysate (LAL) test. Concentrations of airborne endotoxins near barn exhaust fans were significantly higher than at the upwind boundary of the farm and at other farm sites. For most of the study sites, mean concentrations of endotoxins at the downwind boundary of the farm were higher than those at the upwind boundary of the farm, indicating the release of endotoxins from swine CAFOs to the neighboring environment. Endotoxin levels were significantly associated with concentrations of airborne bacteria but not fungi. Environmental factors, such as temperature, relative humidity, and wind velocity, affected the levels of airborne endotoxins at the farms. Based on the ratios of airborne endotoxins in downwind and upwind samples from the farm units, at least five different alternative waste management technologies significantly reduced the release of endotoxins from swine CAFOs. These results suggest that
swine CAFOs are important sources of airborne endotoxins, the levels of which can be reduced by applying more robust and effective waste management technologies.
Introduction Over the last 20 years, the manner of livestock production in many developed countries has dramatically shifted, from traditional crop-livestock farms to industrialized and concentrated livestock production (1, 2). Concentrated animal feeding operations (CAFOs) typically house thousands, sometimes hundreds of thousands, of animals at a single facility. An enormous amount of animal waste is produced by CAFOs, and these byproduct of some animal production systems, such as swine, are typically stored in lagoons on the farm property and eventually applied to agricultural lands (3). During these processes, various microbiological and chemical contaminants can be released, leading to human exposure and contaminating multiple environmental media. Increased numbers of livestock raised in a confined space lead to several potential occupational health and environmental risks (4, 5). For example, various microbiological and chemical components released from CAFOs represent healthrelated risks to workers as well as neighbors in close proximity to animal units (5, 6). Airborne contaminants released from CAFOs include toxic chemicals, infectious agents, and endotoxins. Increased levels of respiratory illnesses, such as infectious diseases, allergies, and asthma, have been documented from exposure to these airborne contaminants on farm properties (7-9). Endotoxins, the lipopolysaccharide components of the outer membrane of Gram-negative bacteria, comprise an important and ubiquitous airborne contaminant found in both urban and rural settings (10). High concentrations of endotoxins have been reported in a variety of settings (9, 11), with prolonged exposure leading to organic dust toxic syndrome as well as several other illnesses (12, 13). Proper treatment and management of animal waste is critical for preventing or minimizing the release of airborne endotoxins. To reduce and better manage the release of environmental contaminants from swine CAFOs, alternative manure treatment and management technologies have recently been developed (14). However, to date, no studies have documented the efficacy of these systems in minimizing the release of airborne endotoxins. To document the efficacy of these alternative systems, levels of endotoxins were measured so that environmental contamination associated with these alternative technologies could be quantified and compared with levels at facilities with conventional waste management systems. The objectives of this study were to evaluate the effectiveness of various animal waste and management technologies in minimizing the amount of airborne endotoxins released from swine CAFOs.
Materials and Methods * Corresponding author e-mail:
[email protected]. Corresponding author address: Department of Environmental Health, Institute of Health and Environment, Seoul National University, Seoul, Korea. † Seoul National University. ‡ The University of North Carolina at Chapel Hill. | Current address: American Chemical Society, 2540 Olentangy River Road Columbus, OH 43210. ¶ Current address: Department of Biological and Agricultural Engineering, North Carolina State University, Raleigh, NC 27695. § North Carolina State University. 3442
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 9, 2010
Sampling Sites. Endotoxin air samples were collected and analyzed from 17 farms located in central and eastern North Carolina. These farms used a variety of waste management technologies: two control farms used conventional technology (lagoon-sprayfield) and 15 additional farms used various alternative waste management and utilization strategies. The number of animals at each farm ranged from 1600 to 7200 animals, except for two farms (farm ID 8, 16) at which the waste treatment technologies were small-scale prototype systems for proof-of-concept application. Entire or partial 10.1021/es9026024
2010 American Chemical Society
Published on Web 03/31/2010
TABLE 1. Summary of Study Farms with Various Manure Treatment Technologies number of animals and type of farm
types of treatments on farma
treatment system open to the environment
farm ID
summary of treatment technology
1 2 3
conventional lagoon (control farm 1) conventional lagoon (control farm 2) solids separation-constructed wetlands system ambient temperature anaerobic digester
4874-head finishing farm 7200-head finishing farm 3520-head finishing farm
B B P-B
entire system open entire system open entire system open
4000-head farrow-to-weana
B
5 6
solids separation-reciprocating wetland solids separation-liquid biofiltration using upflow aerated biological filters
1600-head finishing farm 4000-head finishing farm
P-B P-B
7
4360-head finishing farm
P-B-C
8
solids separation-nitrificationdenitrification/soluble phosphorus removal/solids processing system black soldier fly biocomposting
digester closed, holding ponds open entire system open solids separation open, upflow biological filters closed entire system open
B
entire system enclosed
9 10 11
solids separation-filtramat separator solids separation-screw-press separator high solids anaerobic digester
P-B P-B B
entire system open entire system open anaerobic digester closed
12
B
2700-head finishing farm 9792-head finishing farm
B B-C
6480-head finishing and 1067-head sow farm
B
anaerobic lagoon covered, nitrification and storage pond open entire system open mesophilic digester closed; aerobic digester and water reuse open entire system open
16
permeable anaerobic lagoon with cover, aerated nitrification pond, denitrification/irrigation storage pond sequencing batch reactor mesophilic digester; aerobic digester with methane recovery and power generation, water reuse aerobic blanket-aerated nitrification pond and denitrification/irrigation storage pond gasifier
prototype system servicing less than 40-head finishing barn 4048-head finishing farm 3320-head finishing farm stand-alone facility receiving separated solids from a 4360-head finishing farm 2448-head finishing farm
P
gasifier closed with exhaust standpipe
17
automated windrow composter
prototype system servicing less than 40-head finishing barn stand-alone facility receiving separated solids from a 4360-head finishing farm
B
entire system open
4
13 14 15
a
B, biological; P, physical; P-B, physical-biological; B-C, biological-chemical; P-B-C, physical-biological-chemical.
waste treatment systems were enclosed on many of the farms and were dependent on the types of manure treatment technologies on the farms. The number of air samples collected at each farm ranged from 3 to 28. Table 1 briefly summarizes the characteristics of the farms. Airborne Endotoxin Sampling and Analysis. Airborne endotoxins were measured using 0.8-µm pore size 37-mm PVC filters with cellulose support pads in closed-face cassettes. The sampling was performed using personal SKC air samplers calibrated to sample at 4 LPM for approximately 4 h, typically between 10 a.m. and 5 p.m. In total, 236 airborne endotoxin samples were collected at key sites on the farms between April 2002 and August 2004. For all farms, except farm ID 16, sampling was performed on multiple (2-7) days. Because there was variation in the positions of the waste facilities, “property boundary” data were collected at approximately 150 m upwind (upper boundary) and 150 m downwind (lower boundary) of facilities for consistency when evaluating each technology. At each farm, airborne endotoxin samples were collected near barn exhaust fans (approximately 2 m away from the fan), downwind of lagoon or waste-holding pond edges, at other wastewater treatment unit processes, and at the upwind and downwind boundaries of the farm. Some of the investigated farms did not have a lagoon or separate technology units. For these farms, fewer samples were taken because sampling at the lagoon was not possible. For endotoxin analysis, the endotoxin was eluted from the filter in 10 mL of sterile, pyrogen-free water with 0.05%
Tween-20 at room temperature after 1 h of shaking and centrifugation. The resulting supernatant was transferred to pyrogen-free cryotubes and subsequently assayed. Samples were analyzed by the kinetic chromogenic Limulus amebocyte lysate (LAL) test, as described previously (15). Briefly, the supernatant was diluted serially 2-fold from 1:4 to 1:128. Blank filters were assayed undiluted and at a 1:4 dilution. Collected samples and blanks were assayed on a microplate with a 13-point standard curve using control standard endotoxin. The absorbance was measured on a microplate reader (SpectraMax 340; Molecular Device, Sunnyvale, CA) at 405 nm every 30 s for 90 min. Endotoxin determinations were based on the maximum slope of the absorbance versus time plot for each microplate well compared with the standard curve. Airborne Culturable Bacteria and Fungi. In addition to airborne endotoxin samples, microbiological air samples for culturable bacteria and fungi were collected using AGI-30 impingers (Ace Glass, Vineland, NJ). Impingers were filled with 20 mL of 1% peptone-distilled water (DW) supplemented with 0.01% Tween 80 and 0.005% antifoam A, as described previously (6). Microbiological air samples were collected at 12.5 LPM for 30 min, with flow rates calibrated prior to sampling using a primary flow meter (Dry Cal DC-Lite; BIOS International, Butler, NJ). During microbiological air sampling, the inlet of the sampler (AGI-30) was oriented into the wind. Following sample collection, impingers were immediately stored at 4 °C in the dark for transport to the VOL. 44, NO. 9, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3443
laboratory. Culturable bacteria and fungi were quantified after 10-fold serial dilution using phosphate-buffered saline (pH 7.6) and plating 0.2 mL of each sample in duplicate onto R2A and malt extract agar, respectively. The inoculated plates were cultured at room temperature for 5-7 days for colony development and counting. Environmental Conditions. Environmental conditions, including temperature, relative humidity (RH), wind velocity, and solar irradiation, were recorded simultaneously with air sample collection at a height of 1.5 m above ground level. Temperature and RH were measured using a temperature and RH sensor (model 8720; TSI Inc., Shoreview, MN). Solar irradiation was measured using a radiometer (model 1400A; International Light Inc., Newburyport, MA), and wind velocity and direction were measured using a vane thermoanemometer (Extech Inc., Waltham, MA). During the 4-h sampling, each environmental condition was measured 10 times at multiple sampling locations on each farm. The average of all measurements was calculated and used for statistical analysis. Data Analysis. Bacterial endotoxin concentrations were estimated by dividing the number of EUs from the air filters by the product of the sampling air flow rate and duration of sample collection. Statistical analyses were performed using STATA software (College Station, TX). Before other statistical analysis was performed, the normality of the data was examined. Because normal probability plots indicated that the endotoxin, bacteria, and fungi distributions followed lognormal distributions, these data were log-transformed and used for regression analysis. A t-test was used to compare the concentrations of airborne endotoxins at different sampling sites with those at the upwind boundary on each farm. A Spearman’s rank correlation test was performed to evaluate differences in microbial concentrations under various environmental conditions. Additionally, multivariate regression analysis was used to determine the relationship between airborne endotoxin levels and important independent predictors. Stepwise multivariate regression analysis was performed to investigate microbiological, environmental, and technical factors associated with airborne endotoxin levels on the farms. Endotoxin level (EU/m3) was used as an outcome variable. Predictor variables included total bacteria, total fungi, the number of animals in the barn, type of treatment, system openness, temperature, RH, solar irradiation, and wind velocity. The type of treatment and system
openness were treated as nominal and ordinal categorical variables, respectively. The remaining parameters in the regression were considered continuous variables. In addition to a regression analysis, the ratios of endotoxin levels between the upwind and downwind boundaries were estimated for each farm site sampled. Then, technology performance for each technology was measured by a two sample t-test for identifying better technologies. If endotoxin levels at the downwind boundary were significantly higher than those at the upwind boundary, technology performance was considered lower. If endotoxin levels at the downwind boundary were equal to or lower than those at the upwind boundary, compared with two control farms, technology performance was considered higher.
Results Concentrations of Endotoxins at Different Farms. Concentrations of endotoxins varied significantly across the different farms (Table 2, Figure 1). Overall, the endotoxin concentrations ranged from
