Dennis D. Schulte

Seepage from earthern animal waste ponds and lagoons--an overview of research results and state regulations

Wastewater seepage from earthen animal waste lagoons and storage ponds can contaminate groundwater with nutrients and pathogens. For almost 30 years, the subject has been the focus of laboratory and field research projects designed to (1) measure if and how much earthen ponds and lagoons leak, (2) determine how different soil types affect seepage rates, and (3) evaluate the magnitudes and mechanisms of sealing from animal waste. In this article we present a research review performed to determine how researchers have attempted to answer these questions and how well they have been answered. We discuss weaknesses in the body of knowledge and present further research and educational needs. We also performed a review of 14 state regulations to assess and compare how different states govern seepage from ponds and lagoons. Six states regulate the maximum allowable seepage rate from ponds and lagoons (values ranging from 0.042 to 0.63 cm/day) while another six states regulate the maximum hydraulic conductivity of earthen liners (values ranging from 0.086 to 0.0086 cm/day). The two remaining states regulate neither. The results of this research and regulatory review demonstrate that there is still much to be learned about seepage from animal waste ponds and lagoons. We suggest that a risk-based approach to regulating seepage may be appropriate in the future.

CHARACTERISTICS OF PHOTOTROPHIC AND NON--PHOTOTROPHIC LAGOONS FOR SWINE MANURE

Odors are a major result of inadequately sized and mismanaged anaerobic lagoons. However, purple or pink colored lagoons, indicating the presence of phototrophic purple bacteria, are less likely to be an odor nuisance than are non-purple lagoons. Eight swine lagoons were studied to quantify critical parameters thought to allow purple lagoons to be a more reliable odor control alternative. Bacteriochlorophyll a (Bchl a), which indirectly measures the abundance of phototrophic bacteria, was greater in purple lagoons than in non-purple lagoons (P = 0.01). Oxidation-reduction potential (ORP) was less negative for purple lagoons than for non-purple lagoons in both spring (lagoon temperatures of 6.7C to 8.8C) and during summer (temperatures of 22C to 25C), indicating conditions favoring phototrophism (P = 0.04). Dissolved oxygen levels were near zero and light penetration was minimal in all lagoons. Average sulfide concentrations of all the lagoons were in the range of 1.6 to 6.5 mg/L, which is below the preferred range for purple sulfur bacteria (PSB) growth. Purple lagoons appeared to have lower concentrations of ammonia, alkalinity, chemical oxygen demand, and electrical conductivity among the lagoons studied. Copper and zinc concentrations of all lagoons were not in the range considered to be toxic for anaerobic bacteria. Calculated volatile solids loading rates did not explain differences in Bchl a levels in the lagoons.

TOTAL REDUCED SULFUR CONCENTRATIONS IN THE VICINITY OF BEEF CATTLE FEEDLOTS

A field survey of total reduced sulfur (TRS) concentrations in the vicinity of beef cattle feedlots was conducted to compare field observations against current regulatory thresholds. In addition, environmental factors that may contribute to increased TRS emissions were evaluated. It was observed that TRS levels in the vicinity of beef cattle feedlots are not likely to exceed current regulatory thresholds used by midwestern states. It was further noted that concentration of TRS varies with air temperature and time of day. However, wet feedlot surface conditions and wind speed had almost no impact upon observed concentrations.

SEEPAGE CHARACTERISTICS AND HYDRAULIC PROPERTIES OF A FEEDLOT RUNOFF STORAGE POND

Water and chemical transport were investigated beneath a 22-year-old beef feedlot runoff storage pond. Soil and sludge samples were collected from 14 borings to 6.1-m depths in a cross-section across the pond. The soils consisted of silt loam and clay loam, and the groundwater level was about 30 m beneath the land surface. Soil samples were analyzed for pH, NO3-N, NH4-N, Cl, P, K, SO4-S, TKN, and organic matter. Physical and hydraulic properties were measured on undisturbed samples of soil and sludge to compare saturated hydraulic conductivity, moisture release characteristics, and bulk density. Saturated hydraulic conductivity measured in the laboratory ranged from 0.005 to 0.044 cm/day for the sludge and from 0.008 to 31.4 cm/day for the sidewall soil. The mean hydraulic conductivity values for the sludge and sidewall soil were not significantly different. Mean bulk densities were significantly different. The sludge exhibited high shrinkage when dried, and did not swell to its original volume when rewetted. Moisture content and chemical concentrations were higher beneath the sidewalls than beneath the pond bottom. A seepage rate of 0.87 cm/day was measured after a 7.6-cm rainfall event, following an extended dry period when the pond was empty. This short-term measurement exceeded the allowable seepage rate in Nebraska, however, seepage decreased with time following recharge of the sidewalls. Results have shown that water and chemical movement has occurred beneath the unlined feedlot runoff storage pond and that the plumes have traveled further than the maximum sampled depth of 6.1 m after 22 years of operation.

MODELING SEEPAGE FROM AN UNLINED BEEF CATTLE FEEDLOT RUNOFF STORAGE POND

A site-specific water balance model was developed to evaluate the effects of sludge accumulation, starting stage, and annual precipitation on seepage from an unlined beef cattle feedlot runoff storage pond. The computer model predicted daily inflows due to precipitation and runoff, and outflows due to evaporation and seepage. The seepage component was estimated using the SWMS_2D finite element saturated/unsaturated flow model, while feedlot runoff was estimated using the Natural Resource Conservation Service runoff method. Evaporation, precipitation, and temperature data from a nearby weather station were used in the model. Based on results of 9,100 annual simulations, the mean seepage volume ranged from 31 900 m3/y with no sludge accumulation to 19 300 m3/y with 22 years of sludge accumulation (1.5 m of sludge). The mean seepage rate ranged from 1.11 cm/day with no sludge accumulation to 0.50 cm/day with 22 years of sludge accumulation. Sidewall seepage volumes ranged from 49 to 73% of the total pond seepage volume. Increasing the pond stage from 0 to 250 cm at the beginning of the simulations caused a 200% increase in annual seepage volumes, yet only a 20% increase in annual seepage rates. Annual seepage volumes increased as much as 62% when annual precipitation increased from 44 to 96 cm/y. Average annual seepage rates varied little with varying annual precipitation. Seepage losses were 1.5 to 3.2 times as great as evaporation losses. This research provides information on variability in seepage rates that will be valuable to regulatory personnel when writing new environmental regulations, and to engineers when designing new storage ponds and lagoons.

Phototrophic anaerobic lagoons as affected by copper and zinc in swine diets

Odor emissions from anaerobic lagoons containing large populations of phototrophic bacteria are usually minimal. This study was conducted to determine whether copper (123 ppm) and zinc (2,310 ppm) in diets fed to weanling pigs for therapeutic purposes affect phototrophic conditions within lagoons. Column reactors containing 47 L of swine lagoon sludge and supernatant were used to represent lagoons. The reactors were placed in an environmental chamber maintained at 24° C. Copper, zinc, and control manure were added to the reactors at a volatile solids loading rate of 128 g vs m -3 da-1 using a hydraulic retention time of 32.5 days. Bacteriochlorophyll a, copper, reduction-oxidation potential, salinity, sulfate, sulfide, and zinc were then measured for at least 99 days. Sulfide, total copper and total zinc were the only parameters to be significantly impacted. The copper and zinc concentrations in the sludge increased but that of supernatant in the individual reactors changed little during the study period. However, the addition of dietary copper significantly increased the concentrations of sulfides in the supernatant, creating a condition that appeared toxic to phototrophic bacteria. In contrast, a decrease in sulfide concentration resulted from the addition of dietary zinc, resulting in an environment that may have been favorable to phototrophic bacteria. Thus, to minimize potential odor concerns, zinc rather than copper may be the best choice as a dietary supplement for weanling pigs. Keywords. Anaerobic bacteria, Lagoon effluent, Manure management practices, Odor control, Swine lagoon waste.

Comparison of AERMOD and STINK for Dispersion Modeling of Odorous Compounds

Gaussian dispersion of low-weight volatile fatty acids from a ground-level area source was assessed using STINK (a research-grade, Gaussian plume model from Australia) and AERMOD (an AMS/EPA regulatory model). The goal of this research was to determine if these models could effectively utilize a back-calculation methodology to reasonably assess the emission rate and near-source-dispersion of odorous compounds. VFAs were sampled at six receptors using thermal desorption tubes in one experiment and solid phase microextraction (SPME) fibers in a second experiment. Measured concentrations, when compared with those predicted by STINK and AERMOD, showed good agreement. However, better prediction by AERMOD, compared to STINK, was observed for the experiments involving SPME sampling. The use of AERMOD, combined with the back-calculation approach for the dispersion modeling of odorous constituents, appears promising. The performance of AERMOD was also assessed using meteorological conditions averaged over various time intervals. It was assumed that the average conditions observed during 1, 5, 15 and 30 minute intervals prevailed for the entire 1-hr modeling period. This showed that selection of an appropriate averaging time for meteorological data is important for dispersion modeling when forced to make assumptions about constant meteorological conditions as in the case of Gaussian models like AERMOD.

Ammonia, Hydrogen Sulfide and Odor Emissions from a Beef Cattle Feedlot

Gaseous emissions from livestock facilities represent a concern due to the potential effects of ammonia, hydrogen sulfide, and odors on environmental quality and human health. The lack of knowledge concerning beef cattle feedlot emissions has been a major obstacle in the development of emission reduction strategies and accurate emission modeling. Emissions from a 2,000-head research feedlot in eastern Nebraska were measured in this study using a wind tunnel. Flux rates of ammonia, hydrogen sulfide and odor were determined to compare the effects of diet and pen cleaning frequency for a beef cattle feedlot. Meteorological parameters and soil/manure pH, nitrogen content, surface and subsurface temperature and moisture content were also measured to determine the extent to which they influenced emission rates. Our data clearly show that surface temperature plays an important role in ammonia emissions, but other kinetic factors are also responsible. It also shows that moisture content and temperature influence odor flux, but more research is needed to clarify this relationship. The data suggest diet and cleaning frequency may play a role in ammonia and odor flux.

Modeling Odor Dispersion From a Swine Facility Using AERMOD

Meteorological conditions, odor emissions, and ambient odor levels at a four-barn, swine finishing facility in Iowa were measured in the summer and fall of 2004. This paper compares ambient odor levels measured using a Nasal Ranger® compared to those predicted by AERMOD, a relatively new air dispersion model. Scaling factors needed to adjust predicted odor levels to those observed ranged from 1.66 to 3.12, depending on the source configuration used by the model. Predicted odors levels from the point source configuration required the smallest scaling factor (1.66) and accounted for the greatest percentage of variability in the data when compared to Nasal Ranger readings.

COMPARISON OF AMBIENT ODOR ASSESSMENT TECHNIQUES IN A CONTROLLED ENVIRONMENT

This article compares results of using dynamic triangular forced-choice olfactometry (DTFCO), the Mask Scentometer, the Nasal Ranger, and an odor intensity reference scale (OIRS) to assess odors in a controlled-environment chamber in the Iowa State University Air Dispersion Laboratory. The methods were used to assess 13 odor levels in the chamber. Swine manure mixed with water was used to vary the odor levels. DTFCO did not correlate well to the other ambient odor assessment methods. Predicting dilution to threshold (D/T) using intensity ratings compared to using intensity ratings directly degraded the coefficient of determination (Ro 2 ) through zero with the other methods in all cases. Average intensity-predicted D/T, the Mask Scentometer, and the Nasal Ranger correlated well with each other, with strong Ro 2 values (greater than 0.85) and regression slopes near 1, and the session means were not found to be significantly different ( = 0.05). Using the geometric means of the device D/T settings, (D/T)G, improved the Ro 2 values between the other methods and the Nasal Ranger and Mask Scentometer. Average intensity-predicted D/T values were three to four times higher than Nasal Ranger assessment ((D/T)G and D/T, respectively), and Nasal Ranger (D/T)G was roughly five times higher than Mask Scentometer (D/T)G.