H43B-0371 1340h
Effects of Structural Changes in US Animal Agriculture on Fecal Bacterial Contamination of Streams: Comparison of Confined and Unconfined Livestock Operations
US animal agriculture has undergone major structural changes over the past two decades. Although the total size of the national livestock population (when measured in animal units) has remained relatively unchanged over the period, the number of livestock producers has declined dramatically, and the average size of the remaining operations has increased substantially. A related change has been a pronounced trend towards greater confinement and spatial concentration of farm animals. These changes raise important questions about the water quality effects of animal agriculture. Fewer, larger operations and increased animal confinement might be expected to spatially focus the discharge of manure-related contaminants and result in a smaller number of more heavily impacted watersheds in high production areas. But at a less simplistic level, assessing the overall national effects of the restructuring of animal agriculture requires a spatially detailed analysis covering a wide range of watershed scales. In this study, we use an empirical (SPARROW) model of fluvial fecal coliform bacteria loads to compare the effects of confined and unconfined farm animal populations on levels of fecal contamination in US streams and rivers. The model was calibrated with monitoring data from 341 stream and river monitoring stations distributed among the nation's 48 conterminous states. Bacteria monitoring records cover the period 1978 to 1995. The model accounts for six categories of fecal coliform sources: municipal point sources, urban runoff, forest land, shrublands/wetlands, confined animal agriculture, and unconfined animal agriculture. Bacterial inputs from agricultural sources for the study period are based on data from the Agricultural Statistics Service and the Natural Resources Conservation Service. Model output (fecal coliform loads and concentrations) is produced for 62,000 stream locations.
H43B-0372 1340h
Geochemical Impacts of CAFOs: Implications for Modifying Existing Regulatory Standards in Michigan
Concentrated Animal Feeding Operations (CAFOs), those facilities consisting of 1000 + cattle or hogs, are increasingly among the most significant sources of agricultural contamination in many surface waters in Michigan and around the country. Many environmental and rural citizens' groups argue that Michigan law controlling CAFO operations has been inadequate in dealing with the large volume of nutrient loading observed in streams. Stream impacts due to CAFOs have been measured using the same traditional biological and geochemical parameters as those used to assess general agricultural runoff. Geochemical parameters used include: total nitrogen, nitrate and ammonia, total phosphorous, dissolved oxygen (DO) and conductivity. These parameters are used not only as indicators of stream impact in agricultural states such as Michigan and Indiana but also provide the basis for regulatory standards. This study focused on two large CAFOs in which samples were taken in agricultural drains upstream and downstream from known, historic discharges of these facilities. Traditional analyses of total nitrogen and total phosphorous showed only subtle differences when comparing the upstream and downstream sites. Differences in DO were more pronounced between the sites (consistently lower for the downstream locations) but did not offer a direct correlation with presumed CAFO impacts. Other results, however, suggest that it is possible to isolate CAFO impacts by comparing transient chemical constituents such as (inorganic) orthophosphate and ammonia and by plotting these comparisons against DO. Ammonia concentrations plotted against orthophosphate displayed moderate to strong correlation (r = 0.84) for downstream sites but weak correlations (r = 0.04) for upstream sites. Ammonia and nitrate concentrations correlated strongly for downstream sites (r = 0.58) but weak for upstream sites (r = 0.09). Also, 24 hour DO analyses of all the sites indicated higher fluctuations throughout the day but only for the upstream sites. DO appeared to correlate strongly with orthophosphate concentrations (r = 0.68) for the downstream sites as compared with upstream site correlations (r = 0.058). If new laws and/or guidelines are to be established that adequately address and mitigate potentially negative impacts on surface water from CAFOs, a method of geochemical assessment has to be developed that will isolate CAFO impacts from general agricultural influence. The results of this study suggest that traditional means of assessing agricultural runoff in surface waters are inadequate in determining short and long-term CAFO impacts on surface water quality. Subtle differences do occur using traditional geochemical parameters, but using these parameters alone are insufficient and ineffective in providing a definitive assessment of CAFO impacts relative to the impacts that already exist. What may be needed is a greater focus on the short-term geochemical relationships between parameters such as ammonia, orthophosphate and dissolved oxygen. State and federal guidelines regulating CAFO facilities may need to incorporate these types of analyses in order to better assess future impacts.
H43B-0373 1340h
Denitrification in a Shallow Aquifer Underlying a Dairy Farm in the Central Valley of California
Nitrate loading to shallow aquifers from dairy farm operations presents a serious threat to critical groundwater resources in California. Less well known is the extent to which saturated zone denitrification may mitigate the problem by converting nitrate to the benign end-product nitrogen, before nitrate is transported to deeper aquifers used for drinking water. We are carrying out a multi-disciplinary study of saturated zone denitrification, in a dense network of monitoring points at a 1500-cow dairy in Kings County, California. Detailed vertical profiles of anion and cation concentrations, along with dissolved excess nitrogen were obtained at five-foot intervals using temporary direct-push wells. Results show nitrate concentrations in excess of 100 mg/L over the top few meters of the water column, abruptly falling to less than 5 mg/L below a depth of approximately 10m. Over the same interval, dissolved excess nitrogen concentrations sharply increase, indicating that denitrification is responsible for a significant fraction of the nitrate decrease. This pattern is in effect across the entire dairy site. A key aspect of the project is a concurrent focus on understanding the hydrogeology of the site. Regionally, overdraft over the past several decades has resulted in the development of separate shallow (10 m) and deeper ($\ge$ 40 m) aquifer systems. Recharge to the shallow aquifer is derived from low TDS, isotopically depleted Kings River water from a nearby unlined irrigation canal. Local agricultural pumping from the shallow aquifer and infiltration from irrigation water are significant factors in the shallow system. The deeper aquifer is characterized by intensive regional pumping, rapidly decreasing water levels, and the apparent disposition of the shallow aquifer as a perched system for a 1-km$^2$ or more area surrounding the farm. The air gap separating the aquifers is low in oxygen and undergoes pressure changes as water levels fluctuate below. Age dating and negligible nitrate levels in the lower aquifer suggest a long horizontal recharge pathway from the perimeter of the perched area, while sustainability of the shallow system is dependent upon continuing recharge through the irrigation canal. This work was conducted under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under contract W-7405-Eng-48.
H43B-0374 1340h
Dairy Wastewater, Aquaculture, and Spawning Fish as Sources of Steroid Hormones in the Aquatic Environment
A suite of androgens, estrogens, and progestins were measured in samples from dairy farms, aquaculture facilities, and surface waters with actively spawning fish using gas chromatography-tandem mass spectrometry (GC/MS/MS) to assess the potential importance of these sources of steroid hormones to surface waters. In a dairy waste lagoon, the endogenous estrogens 17beta-estradiol and estrone, and the androgens testosterone and androstenedione were detected at concentrations as high as 650 ng/L. Samples from nearby groundwater monitoring wells demonstrated removal of steroid hormones in the subsurface. Samples from nearby surface waters and tile drains likely impacted by animal wastes demonstrated the sporadic presence of the steroids 17beta-estradiol, estrone, testosterone, and medroxyprogesterone, usually at concentrations near or below 1 ng/L. The endogenous steroids estrone, testosterone, and androstenedione were detected in the raceways and effluents of three fish hatcheries at concentrations near 1 ng/L. Similar concentrations were detected in a river containing spawning adult Chinook salmon. These results indicate that dairy wastewater, aquaculture effluents, and even spawning fish are sources that can lead to detectable concentrations of steroid hormones in surface waters and that the concentrations of these compounds exhibit considerable temporal and spatial variation.
H43B-0375 1340h
Transport of microorganisms in the presence and absence of manure suspensions
Wash water and storm water runoff from Concentrated Animal Feeding Operations (CAFOs) frequently contain manure and a variety of viral, bacterial, and protozoan parasite pathogens. Column experiments were conducted to elucidate the transport behavior of representative microbes (coliphage, Escherichia coli O157:H7, and Giardia cysts) through several aquifer sands in the presence and absence of manure suspensions. Specific factors that were considered include the soil grain size distribution, the presence and absence of manure suspensions, and manure size distribution. Effluent concentration curves and the final spatial distributions of microorganisms and manure particles were measured. Increasing the microbe size and decreasing the median grain size of the sand resulted in low effluent concentrations and increased retention of the microbes, especially in the sand near the column inlet. Similar transport trends were observed for the manure suspensions in these sands. The spatial distributions of retained microbes and manure were generally not consistent with predictions from conventional attachment, detachment, and blocking models; but rather with straining. The transport potential of the microbes was sometimes enhanced in the presence of manure suspensions. This observation, as well transport and retention data for manure suspensions, suggest that manure components filled straining sites and inhibited microbe retention. Differences in the surface charge properties of clean and manure equilibrated microbes (presumably due to adsorption of organic components from the suspension) may also influence transport behavior.
H43B-0376 1340h
Development and testing of real-time PCR assays for determining fecal loading and source identification (cattle, human, etc.) in surface water and groundwater
A multi-disciplinary group of researchers at the University of Tennessee is developing and testing a series of microbial assay methods based on real-time PCR to detect fecal bacterial concentrations and host sources in water samples. Real-time PCR is an enumeration technique based on the unique and conserved nucleic acid sequences present in all organisms. The first research task was development of an assay (AllBac) to detect total amount of Bacteroides, which represents up to 30 percent of fecal mass. Subsequent assays were developed to detect Bacteroides from cattle (BoBac) and humans (HuBac) using 16sRNA genes based on DNA sequences in the national GenBank, as well as sequences from local fecal samples. The assays potentially have significant advantages over conventional bacterial source tracking methods because: 1. unlike traditional enumeration methods, they do not require bacterial cultivation; 2. there are no known non-fecal sources of Bacteroides; 3. the assays are quantitative with results for total concentration and for each species expressed in mg/l; and 4. they show little regional variation within host species, meaning that they do not require development of extensive local gene libraries. The AllBac and BoBac assays have been used in a study of fecal contamination in a small rural watershed (Stock Creek) near Knoxville, TN, and have proven useful in identification of areas where cattle represent a significant fecal input and in development of BMPs. It is expected that these types of assays (and future assays for birds, hogs, etc.) could have broad applications in monitoring fecal impacts from Animal Feeding Operations, as well as from wildlife and human sources.