Full
Report PDF (5.2 MB) |
This report presents the results of a major research effort to investigate technologies and approaches to reducing emissions of odor, gases (ammonia and hydrogen sulfide), and dust from commercial swine operations in Illinois. The research was carried out by the University of Illinois at the Henco Hogs facility near Monmouth, IL, a typical modern farrow-to-wean production facility between August, 2002 and December, 2003.
A total of 15 studies were completed and a review of other potential technologies that are currently available for application on swine facilities to manage manure and control emissions was carried out. These are summarized below.
A preliminary study carried out to determine the effect of room and sampling day on odor and gas levels in the farrowing facility in September and October of 2002. Four rooms of a 14-room complex were studied in two replicates over time with samples being collected on days 1, 8, and 15. There was an effect of sampling day and farrowing room on concentrations of odor, H 2S and NH 3. Gas concentrations increased linearly with sampling day (time sows were in room) and were generally similar in the rooms in the center of the building but different between the rooms in the center compared to those at the ends of the building. These results were used in the design of subsequent studies.
Six rooms of the 19-room farrowing complex were studied for a two-year period from September 2002 through August 2004. Rooms were occupied for 15-day periods by a series of groups of sows. Samples were collected on days 1, 8, and 15 of each period. Over the period of the study, odor units ranged from 2.02 to 3.46, averaging 2.81. Dilution threshold ranged from 104 to 2866, averaging 778. NH 3 values ranged from non-detectable to 7.20 ppm, H 2S from non-detectable to 2.40 ppm. Average values of the 24-mo period were 1.97 ppm for NH 3, 0.40 ppm for H 2S. The regressions of gas levels against time suggested either little change (for H 2S), or a small reduction in concentration (for odor and NH 3) over the two-year study period.
Sampling (at the room exhaust and pit exhaust fans) was carried out over a two-year period from September 2002 through August 2004. Odor units ranged from 2.50 to 3.26, with an average of 2.79, and dilution threshold from 358 to 2140, with an average of 725. NH 3 ranged from non-detectable to 23.10 ppm. H 2 S from non-detectable to 0.2 ppm. Average values for the 24-mo period were 8.96 ppm for NH 3 and 0.01 ppm for H 2S. The regressions of gas levels against time suggested a significant reduction in concentrations (for H 2S odor and NH 3) over the two-year study period.
Two treatments were compared : control and flushed. The study period was 15 d. Pits were flushed before the study started for the control treatment only and before the start and on study d 8 for the flush treatment. On d 8, all odor and gas values were numerically lower after flushing the pit than before but only H 2 S differed significantly (P<0.05). H 2 S measured 4 h before and 4 h after flushing was 0.54 and 0.10 ppm ( SEM 0.119), respectively, in room exhaust, 0.87 and 0.24 ppm ( SEM 0.125) in pit exhaust. However, on d 15 (1 wk after flushing) odor and gas levels were similar in control and flushed rooms. These results suggest that flushing pits had a limited, short-term effect on odor, NH 3 , and H 2 S in the air exhausted from the farrowing rooms.
Puremax water treatment passes pass a current through the liquid manure in the pit and is claimed to reduce mineral build-up and increase biological activity in stored manure. Two rooms of a 14-room farrowing complex were used in three replicates over time. Two treatments were compared : control (untreated) and Puremax. One room served as the control, the other was equipped with the Puremax system. There was no effect of the treatment on odor, NH 3, or H 2S at the pit exhaust fan, however, ammonia concentration was lower at the pit exhaust fan for the Puremax-treated room. These r esults suggest that the Puremax system lowered NH 3 at the pit exhaust fan.
Two rooms of a 14-room farrowing complex were evaluated with one being used as the control (untreated) and the other being treated (i.e., oil sprinkled daily across all surfaces). Major problems were experienced in the oil-treated rooms with animal mobility and health and worker comfort and safety that resulted in the premature termination of this study. The use of oil sprinkling in this or any other farrowing facility is inappropriate.
Two lactation dietary fat treatments were compared : control (3.5% fat) and a high-fat diet (6.5% fat) (the standard diet used on this facility). Two rooms of a 19-room farrowing complex were used in four replicates over time. There were no significant differences in odor, NH 3, or H 2S at room exhaust or pit exhaust fan. Dust mass and particle size did not differ (P>0.05) between rooms. There is no basis from these results for recommending a change in the current practice of including 6.5% fat in the lactation diets fed in the farrowing facility.
Two lactation dietary protein treatments were compared : control (21.2% dietary crude protein) and low-protein (19.7%). Two rooms of a 19-room farrowing complex were used in three replicates over time. The higher crude protein level (control diet) was the standard lactation diet at this facility, the lower was considered the lowest that would not affect sow performance. There were no significant treatment effects on odor, NH 3, or H 2S at room exhaust or pit exhaust fan. Dust mass and particle size did not differ (P>0.05) between treatments. There is no basis from these results for recommending a change in the current practice of including 21.2% crude protein in the lactation diets fed in the farrowing facility.
The air dam was constructed to cover all 14 of the room exhaust fans on the South end of the gestation barn. The Biocurtain was placed on the North side of the gestation house, covering all 13 of the room exhaust fans and on the east side of the farrowing unit. The Biocurtain constructed on the outside of the farrowing facility covered room exhaust and pit exhaust fans of the initial 14 farrowing rooms (before expansion of the farrowing facility). Samples were taken every two weeks during a three-month period (June – August 2004). For the air dam, samples were taken directly inside and directly outside the room exhaust fan, before and after the air dam and at three points at the top of the dam. Samples for the Biocurtain fitted to the farrowing and gestation building were taken at the same locations (only one sample was taken at the top of the structure). The new farrowing rooms without the Biocurtain were used as a control for comparison with the samples taken on the farrowing rooms fitted with the Biocurtain; samples were taken at the same locations as for those fitted with the Biocurtain. Results showed no clear pattern with respect to the effect of either the air dam or the Biocurtain on odor levels. Concentrations of NH 3 were similar before and after the Biocurtain, however, the airflow rate was higher through the opening at the top of the Biocurtain which suggests that the majority of the NH 3 escaped the Biocurtain through the top opening. H 2S could not be determined due to detection limits of the sampling tubes. Dust measurements showed variable effects of the experimental treatments on the size of the particles leaving the air dam and Biocurtain. On the basis of this limited evaluation, it is not possible to conclude if the air dam or the Biocurtain had any positive influence on emissions from this facility.
The biofilter was built as two separate and operationally distinct parallel units, each having its own air supply fan from the manure tank headspace. Two biofilter media were used : (1) chipped bark (“debark”) and (2) hay/wood chips mixture. Moisture addition was via soaker hoses that were buried in the biofilter medium during initial medium loading. Moisture was monitored using a novel approach. Design operating parameters of the biofilter were a minimum empty-bed contact time of 10 sec, a uniform airflow rate of ~6 ft/min, and a pressure drop through the biofilter of ~0.15 inches H 2O. NH 3 reduction by the biofilter averaged 24% for medium 1, 67% for medium 2. H 2S in air coming from the manure tank was almost below the range of detection throughout the experiment for the instrument used (colorimeter tubes). The biofilter reduced odor concentrations from an inlet average (manure tank exhaust) of 1005 odor units down to an average 323 for medium 1, from 1184 odor units to 300 for medium 2. The biofilter proved effective at reducing odorous and NH 3 emissions from the manure storage tank.
A vertical diffusion-coagulation-separation ( DCS) deduster was constructed and installed on two pit exhaust fans of the breeding and gestation building. The concentrations of dust and gases were measured at two test conditions : (1) dry condition or no water sprays in the diffusion-coagulation section, and (2) wet condition or when water was sprayed on the incoming air for two minutes of every hour. There was no measurable change in odor concentration measured between the inlet and outlet ports of the deduster. The removal efficiency for total suspended particles and gases was higher under the wet (with water spray) than under the dry (no water spray) conditions. The average removal efficiencies for dust were about 52 and 59% and for NH 3 were 13 and 15% for dry and wet conditions, respectively. The deduster was not effective in reducing odor from the swine building but was effective at reducing ammonia and, particularly, dust emissions.
The objective was to develop an approach to monitor the volume of air being exhausted from the fans in the farrowing and gestation facility. The primary measurement system was based on a vane anemometer fitted to each exhaust fan linked to a data logger. Back up systems were based on a static pressure sensor that measured the pressure difference between the room and the ambient condition and a voltage transducer that monitored the power supply to the fans.
Model simulations of odor transport provide a means of estimating odor levels downwind of a confinement facility. The odor emanating from the Henco facility was characterized based on the measurements of odor and ventilation rates taken in the study. Baseline odor levels were estimated assuming no treatments were in place. Baseline levels then provide a basis for evaluating effects of air dam/Biocurtains and biofilters on odor levels. Simulations showed very little odor reduction associated with the air dam/Biocurtain. Although both of these options release odor at an elevated height, this was not sufficient to influence downwind odor dispersion. Conversely, simulations of emitting odor at greater heights of 50 and 100 ft show substantially lower downwind odor levels. Assuming that a biofilter operated at 70% efficiency and that all the building air was so treated would result in a substantial reduction of odor pattern. In this example, the area over which the odor is classified as “faint” would be reduced from >1 mile to ~¾ mile.
The economics of operating various technologies and approaches evaluated in this research was calculated.
- Study IV. Pit Flushing Frequency :Total cost per flush equaled $13.75 ($3.75 for labor, $10 for utilities).
- Study V. Puremax System : Total costs were $19.25 for the 15 day analysis period ($10.85 fixed costs plus $8.40 electricity costs).
- Study VI. Oil Sprinkling : The total cost of this treatment over 15 d was $89.33
($33.08 of oil = 10.5 gallons x $3.15, $56.25 of labor = 3.75 x $15).- Study VII. High Fat Lactation Diets : The high fat diet cost $2.66 more than the low fat diet.
- Study VIII. Low Crude Protein Lactation Diets : The low protein diet cost $1.62 more than the high protein diet.
- Study X. Biofilter : Total cost for operating the biofilter was $972.18/year, comprised of $202.67 for concrete structure, $135.98 for concrete slats, $175.97 for fan, $107.56 for biomedium, and $350 for electricity.
- Study XI. Diffusion-Coagulation-Separation Deduster : Total cost per flush for each room was $13.75 ($3.75 for labor, $10 for utilities).
Neighbors were interviewed four times at roughly 6-month intervals throughout the 2-year study period, the first interview taking place spring 2003, ~6 mo after testing started. Interviews also were conducted in fall 2003, spring 2004, and fall 2004. In the earlier interviews, there were mixed responses regarding changes in odor levels from the facility (three neighbors felt there was no change, two felt it had declined slightly, two felt it had increased slightly). However, at the final interview, all neighbors interviewed indicated that the odor had declined over the last 6 mo of the project (half of the neighbors felt that the odor had declined greatly, and the remainder felt that it had declined slightly). Neighbor responses at the end of the study period were very positive regarding the decrease in odors from this facility.
A search of literature and websites relating to manure management and emission reduction was carried out and a data base was constructed summarizing technologies and approaches, including those not evaluated in this research project.
Full
Report PDF (5.2 MB) |
