Surface water and groundwater quality impacts at a swine CAFO with a capped lagoon system in eastern North Carolina By Lauren Richardson May, 2023 Co-Directors of Thesis: Guy Iverson and Stephen Moysey Major Department: Geological Sciences ABSTRACT Waste from concentrated animal feeding operations (CAFOs) is traditionally stored in open lagoons and the liquid wastewater is periodically applied to nearby sprayfields. Capping lagoons with an impermeable cover have been proposed to reduce environmental impacts by inhibiting the release of odors and accidental overflows, but there is a lack of data evaluating how capping a lagoon affects surface water and groundwater quality near the sprayfields. This study analyzed concentrations and masses of nitrogen in surface water downgradient of sprayfields compared to upstream reaches. Furthermore, this study investigates nitrogen concentrations in groundwater beneath and downgradient from sprayfields compared to background groundwater. Water samples were collected monthly to quantify concentrations and/or masses of total dissolved nitrogen (TDN) and other dissolved nitrogen species between October 2019 to January 2021. Results indicated that surface water downstream of the farm contained a median TDN concentration that was 14 times greater (6.65 mg L-1) than upstream (0.48 mg L-1). Median flux of TDN downstream of the farm (32.1 kg ha1 yr-1) was 11 times greater and statistically different than upstream (2.8 kg ha1 yr-1). In-situ, high frequency monitoring performed over 2 weeks (8/14/22-8/31/22) showed and elevated concentration and range of TDN downstream of the site (1.9-137.9 mg L-1) compared to upstream (0.1-4.8 mg L-1), which was attributed to baseflow inputs and spray events that occurred 6 days after deployment. Increases in surfaced water TDN were likely from wastewater applications to sprayfields. The lagoon wastewater had a median TDN concentration of 485 mg/L and TDN concentration reductions of 94.4-99.6% were observed between the lagoon and the nearby groundwater. The median TDN concentrations of groundwater under sprayfields (5.0 mg L-1) was 12 times higher than background (0.42 mg L-1). Groundwater beneath the riparian buffers was also elevated (6.4 mg L-1) and 15 times higher than the background groundwater. Despite major reductions between the lagoon waste and the sprayfields where the waste is applied, TDN concentration in groundwater remained elevated compared to background levels. These data suggest that land application of wastewater contributed to elevated TDN concentrations in water resources downgradient of the farm. Strategies to ameliorate nitrogen in subsurface and surface discharge from the CAFO may be needed to reduce net nitrogen exports and impacts from the farm. Surface water and groundwater quality impacts at a swine CAFO with a capped lagoon system in Eastern North Carolina A Thesis Presented to the Faculty of the Department of Department of Geological Sciences East Carolina University In Partial Fulfillment of the Requirements for the Degree Master of Science in Geology By Lauren Richardson May, 2023 Director of Thesis: Guy Iverson, Ph.D.; Stephen Moysey, Ph.D. Thesis Committee Members: Charles Humphrey, Ph.D. Michael O’Driscoll, Ph.D. © Lauren Richardson, 2023 Surface water and groundwater quality impacts at a swine CAFO with a capped lagoon system in Eastern North Carolina By Lauren Richardson APPROVED BY: Co-Director of Thesis Guy Iverson, Ph.D. Co-Director of Thesis Stephen Moysey, Ph.D. Committee Member Charles Humphrey, Ph.D. Committee Member Michael O’Driscoll, Ph.D. Chair of Geological Sciences Eric Horsman, Ph.D. Interim Dean of the Graduate School Kathleen T Cox, Ph.D. AKNOWLEDGEMENTS I would like to thank the UNC Inter-Institutional Planning Grant “Building Team Science to Support Sustainable Animal Agriculture in North Carolina” for providing us funding and support for this research study. A very special thank you to my co directors, Dr. Stephen Moysey and Dr. Guy Iverson. You all pushed me to be better and guided me when I needed it the most. Thank you to my other committee members, Dr. Charles Humphrey and Dr. O’Driscoll, for all your guidance and support throughout my thesis research and my time spent in your classrooms. To my entire committee, I would not be where I am today if it weren’t for you all and all the opportunities you provided. Thank you to the entire East Carolina University Geological Sciences Department and the Department of Health Education & Promotion for creating a diverse and interdisciplinary experience I had over the course of my research and time as a student. Furthermore, thank you to Dr. John Hoben for all the roles you played; the best field partner, patient teacher, and friend. Thank you to Melissa Nolan for all your help in the field and the lab, but more importantly, thank you for your friendship and positivity. Thank you to my other helping hands in the field and the lab; Bobby Vaughan, AJ Discepolo, and John Willis. Thesis data would not have been possible without Suelen Tullio and the other wonderful staff at the Environmental Research Laboratory. A special thank you to the property owner where we conducted this study. Your willingness to adapt and allow us on your farm is inspiring. I hope we as scientists and those in the industry will continue to grow our professional relationships and work together to provide a better future for the state of North Carolina and all who live here. Finally, thank you to my greatest support and biggest cheerleader, Garrett Elmo. At the beginning of this project, you were my best friend, but now I get to call you my husband. Your love and understanding have gotten me to the finish line. Table of Contents LIST OF TABLES ............................................................................................................................................ vii LIST OF FIGURES ......................................................................................................................................... viii CHAPTER 1: INTRODUCTION ........................................................................................................................ 1 1.1 Water Quality Issues ........................................................................................................................... 1 1.2 Agriculture Farm Operations .............................................................................................................. 2 1.3 Fate and Transport of Nitrogen from Animal Agriculture .................................................................. 4 1.4. A Need for Alternative Waste Management? ................................................................................... 7 1.5 Organization and Study Objectives ..................................................................................................... 8 CHAPTER 2: QUANTIFYING NITROGEN TRANSPORT TO SURFACE WATERS DOWNGRADIENT OF A SWINE CAFO IN EASTERN NORTH .............................................................................................................. 10 2.1 Introduction ...................................................................................................................................... 10 2.2 Methodology ..................................................................................................................................... 13 2.2.1 Study Area .................................................................................................................................. 13 2.2.2 Instrumentation of Surface Water Sites .................................................................................... 16 2.2.3 Sampling Frequency and Protocols ............................................................................................ 17 2.2.4 Statistical Analysis ...................................................................................................................... 18 2.3 Results and Discussion ...................................................................................................................... 19 2.3.1 Nitrogen Concentrations ............................................................................................................ 19 2.3.2 Sources and Pathways of TDN ................................................................................................... 27 2.3.3 Surface Water Discharge and Exports........................................................................................ 31 2.4 Conclusion ......................................................................................................................................... 38 CHAPTER 3: ANIMAL WASTEWATER MANAGEMENT IMPLICATIONS BASED ON GROUNDWATER NITROGEN CONCENTRATION AND EXPORT IMPACTS AT A SWINE CAFO IN EASTERN NORTH CAROLINA .................................................................................................................................................................... 41 3.1 Introduction ...................................................................................................................................... 41 3.2 Methodology ..................................................................................................................................... 47 3.2.1 Study Area .................................................................................................................................. 47 3.2.2 Installation of Groundwater Piezometers .................................................................................. 49 3.2.3 Sampling Frequency and Protocols ............................................................................................ 50 3.2.4 Data Analysis .............................................................................................................................. 52 3.3 Results and Discussion ...................................................................................................................... 54 3.3.1. Hydrology .................................................................................................................................. 54 3.3. Specific Conductivity and Geophysical Data ................................................................................ 61 3.3.2 Nitrogen Concentrations ............................................................................................................ 63 3.3.3. Mass Reduction Estimates ........................................................................................................ 68 3.3.4 Sources and Pathways of TDN ................................................................................................... 71 3.4 Conclusion ......................................................................................................................................... 73 CHAPTER 4: CONCLUSION AND MANAGEMENT IMPLICATIONS .............................................................. 74 4.1. Groundwater-Surface Water Interactions ....................................................................................... 74 5.2 Management Implications ................................................................................................................ 76 REFERENCES ................................................................................................................................................ 78 APPENDIX A: SUMMARY OF LAND COVER .................................................................................................. 89 APPENDIX B: SOIL SERIES SUMMARY .......................................................................................................... 91 APPENDIX C: SURFACE WATER STATISTICS ................................................................................................. 92 APPENDIX D: RAW QUALITY AND QUANTITY DATA .................................................................................... 93 APPENDIX E: PRECIPITATION DATA FROM HIGH FREQUENCY EVENT ...................................................... 106 APPENDIX F: SPECIFIC CONDUCTIVITY FIELD EVENT ................................................................................ 107 APPENDIX G: SEEP-UP AND SEEP-DOWN WATERSHED ESTIMATE ........................................................... 108 APPENDIX H: GROUNDWATER STATISTICAL SUMMRY ............................................................................. 109 APPENDIX I: MIXING MODEL RANGE ESTIMATES ..................................................................................... 112 LIST OF TABLES 1 Summary of watershed area, stream gradient, proximity to sprayfield, and soil series 16 2 Sprayfield irrigation by date during high frequency event 27 3 Summary of physical and chemical parameters at surface water locations 30 4 Summary of concentrations and exports in surface water 36 5 Summary of literature findings 46 6 Summary of soil type, elevation, and depth for groundwater locations 48 7 Summary of physical and chemical parameters in groundwater 57 8 Summary of concentration data for groundwater locations 66 9 Mixing model 70 LIST OF FIGURES 1 Map of study area and surface water sampling locations 15 2 Surface water nitrogen concentration box plots 20 3 Surface water nitrogen concentrations time series 21 4 Surface water nitrogen high frequency data time series before and during a spray event 26 5 Surface water isotope data 28 6 Map of conductivity measurements in stream and seep 31 7 Normalized discharge for upstream and downstream 33 8 Precipitation data and temperature data for study period 34 9 Total dissolved nitrogen exports in surface water 37 10 Study area map and groundwater sampling locations 49 11 Monthly precipitation data and temperature of groundwater 55 12 Depth to water for groundwater piezometers 58 13 Time series of depth to water for piezometers 59 14 Average groundwater flow direction 61 15 Specific conductivity for groundwater locations 62 16 OhmMapper and resistivity data for sprayfield 63 17 Nitrogen concentrations for groundwater locations 67 18 Isotope date for groundwater 72 CHAPTER 1: INTRODUCTION 1.1 Water Quality Issues Nitrogen is a life-essential nutrient which often limits primary productivity in ecosystems where it Is not readily available. However, elevated amounts can lead to negative environmental and human health outcomes especially in sensitive ecosystems like coastal areas. Increased availability of plant nutrients (e.g., nitrogen) is a major issue in coastal areas. Too much nitrogen can cause negative environmental impacts like eutrophication, algal blooms, and fish kills (Vitousek et al., 1997; Tilman et al., 2002; Townsend et al., 2003). Human health impacts like methemoglobinemia and cancer have been associated with increased nitrate consumption, leading to the establishment of a drinking water standard for nitrate of 10 mg/L based on these health concerns (Sadeq et al., 2008). Drinking water standards do not necessarily protect against environmental impacts, however, since eutrophication can occur at lower concentrations (Osmond et al., 2003; Humphrey et al., 2010). Onsite wastewater treatment, municipal waste treatment plants, industrial facilities, atmospheric deposition, fertilizers, agriculture, and stormwater runoff are all common sources of nitrogen species (Paerl et al., 1995; Castro et al., 2003; Mallin and Cahoon, 2003; Humphrey et al., 2016; Mallin et al., 2015). Of these, agriculture has been cited as the largest nonpoint source contributor to water quality issues and degradation for the Cape Fear, Neuse, and Tar-Pamlico River Basins and their associated estuaries and lakes (NC DEQ, 2018). Excess nitrogen is a concern in the Coastal Plain of North Carolina specifically because of the large volumes of wastewater produced at concentrated animal feeding operations (CAFOs) that can degrade the environment (Stone et al., 1998a). According to Martin et al. (2018) over 55% of CAFOs in North Carolina are located in the Coastal Plain. The Coastal Plain is low lying and vulnerable to storm events and flooding which can lead to nutrients being transported to nearby surface water (Martin et al., 2018). 1.2 Agriculture Farm Operations North Carolina agriculture operations are key producers of crops and livestock. Row-crop agriculture is focused on planting, harvesting, and re-planting crops on a large scale. Row-crop operations use best management practices to prepare the soil for planting, irrigate appropriately, and apply fertilizer to produce high yields efficiently (Roberts, 2007). Crop production is primarily nitrogen limited, thus nitrogen fertilizers are applied to agricultural fields to increase production (Roberts, 2007). Irrigation is necessary for crop health, but irrigative practices must be timed appropriately with fertilizer application and local weather to mitigate nitrate leaching and/or runoff. Irrigation management is also important because excessively irrigating crops may reduce the efficiency of the fertilizer applied, therefore decreasing the likelihood of a profitable harvest (Roberts, 2007). North Carolina is also a major producer for animal agriculture, specifically swine CAFOs. A CAFO is an individual concentrated feeding operation that house large amounts of animals in a confined structure to meet production needs. Instead of living in fields or pastures, the animals lived, eat, and create waste in a confined structure (Mallin and Cahoon, 2003). A typical swine CAFO consists of four main components, including: hog houses to store the animals, lagoons where the hogs’ waste is stored, sprayfields where the waste is applied, and setback distances between sprayfields and water resources. Typically, animal waste is generated in the hog houses and then transported to waste lagoons or pits where it is stored and treated 2 anaerobically according to the CAFO’s certified animal waste management plan (Harden, 2015; Ducey et al., 2019). A National Pollutant Discharge Elimination System permit may be required if the CAFO exceeds 2,500 swine that weigh more than 55 pounds (U.S. Code § 1251). The waste in the lagoon is kept at a pH above 7 through liming, which promotes formation of ammonia (Mallin and Cahoon, 2003; NC General Statute § 143-215.10C). The EPA has separation requirements to address nutrient management and odor control regarding animal agriculture waste. Most swine CAFOs in North Carolina house at least 2,500 hogs and rely on large waste lagoons (e.g., 0.5 to 2 acres and 2.4 to 6.0 m deep) to manage wastewater generated by livestock (Mallin et al., 2015). Traditionally, wastewater management includes periodic land application to supply nearby sprayfields with nitrogen and other nutrients to support the growth of crops. Land application of waste is limited to the growing season, March-September, and must be applied at times that will result in realistic production goals to minimize excess nitrogen nutrients to surface water and groundwater (US EPA, 2003). Wastewater application rates are based on nutrient management plans identifying the concentration of plant available nitrogen (PAN) that can be utilized by crops in the sprayfields. This is to mitigate over-application of nutrients, which, if not used by plants, may leach or runoff to downgradient water resources (Christenson and Serre, 2017). Sprayfields receiving waste require 30 m (100 ft) horizontal distance from surface water. The 30 m setback distance is not required if the farmer or operator substitutes a vegetated 10 m (35 ft) buffer (US EPA, 2012). In North Carolina, a moratorium was implemented on new and expanded swine farms in 1997 and then made permanent in 2007 for swine farms using anaerobic waste lagoons. Any new farms or farms that choose to expand must meet stricter environmental quality standards, such as 3 constructing lagoons with synthetic liners and eliminating odor emissions beyond the farm parcel boundaries. Existing farms can continue to operate without necessarily meeting the new standards if the farm does not expand (NC General Statute § 15A NCAC 02T .1307-.1308 and 15A NCAC 02D .1808). Montefiore et al. (2022) used Landsat imagery of lagoons to determine construction timelines and found that new construction of waste lagoons decreased dramatically after the moratorium was implemented. 1.3 Fate and Transport of Nitrogen from Animal Agriculture Industrialization of small-scale, “traditional” animal farms has resulted in large, commercialized farms housing hundreds to thousands of animals. These industrial farms, or CAFOs, mass produce animal products at lower costs (Brown et al., 2020). CAFOs are a major source of labile nitrogen to waterways primarily because of how the animal waste is stored and managed at these sites. The waste is predominantly ammonium and organic nitrogen (Burns et al., 1987; Harden, 2015). Waste from the lagoon is periodically extracted and applied onto nearby sprayfields for plant uptake based on a waste management plan (Spruill et al., 2005; Harden, 2015). The ammonium-rich waste applied to the sprayfields may undergo nitrification if aerobic conditions are present in the subsoil beneath the application field (Behnke, 1975; Humphrey et al., 2010; Del Rosario et al., 2014). Transformations from ammonium to nitrate (nitrification) require an aerobic environment, a carbon source, nitrifying bacteria, and the presence of ammonium. Nitrate is prone to leaching (Behnke, 1975; Wilhelm et al., 1996; Kester et al., 1997; Qiao et al., 2015) and may result in elevated nitrogen concentrations in groundwater beneath sprayfields that can contribute to environmental health issues. Ammonium may also be transported to groundwater according to some studies (Bouwer et al., 1980). If ammonium is 4 applied to soil where cation exchange sites are full, the ammonium can leach to groundwater (Wilhelm et al., 1996). Nitrate reductions may occur in groundwater due to denitrification or plant uptake (Stone et al., 1998b). Denitrification requires anerobic conditions, a carbon source, denitrifying bacteria, and the presence of nitrate (Wilhelm et al., 1996). A common limitation that inhibits denitrification in groundwater systems is the availability of electron donors (DeSimone and Howes, 1996). Agricultural waste management practices have been linked to excess nitrogen species in sensitive watersheds (Harden, 2015). Nutrients from waste applied to sprayfields for the purpose of plant uptake has been found in downgradient groundwater and surface water resources (Israel et al., 2005; Harden, 2015). Factors that can increase potential for runoff and leaching of excess nitrogen are climate and soil type (Wick et al., 2012). Climate can increase leaching if the precipitation exceeds evapotranspiration (Donner et al., 2004). Soil type is also a factor in leaching because coarser grains, like sandy soils found in coastal areas, may allow more nitrogen to leach into groundwater since these soil types have lower capacity to hold water or nutrients (Steenvorden et al., 1986). Sandy soils have interconnected pores that increase effective porosity, therefore increasing permeability. Clay soils are porous but lack interconnected pores which will impede or restrict flow (Heath, 1984). Past studies (Mallin et al., 2004; Harden, 2015) found animal agriculture waste management is a source of excess nitrogen species in surface water nearby CAFOs. For example, Harden (2015) quantified nitrogen loading from 54 stream sites draining agricultural lands in multiple river basins in North Carolina (including Tar-Pamlico, Neuse, and Cape Fear). Harden (2015) found that catchments containing swine farms contained mass loads of nitrogen that were 1.5x greater than background (farms with no animal agriculture). Mallin et al. (2015) sampled 5 surface water from the Stocking Head Creek Watershed, which, at the time of the study, contained 13 swine CAFOs and 11 poultry CAFOs. They study found elevated ammonium levels from 0.5 mg/L to 37.8 mg/L and total nitrogen ranging from 0.11 mg/L to 46.70 mg/L. Both Mallin et al. (2004) and Harden (2015) concluded that watersheds containing CAFOs may contribute to water quality degradation that can cause human and environmental impacts. In addition to surface water impacts, past studies have found that CAFOs can negatively affect groundwater quality. Karr et al. (2001) and Israel et al. (2005) have also found that groundwater underneath and nearby sprayfields may contain elevated nitrogen species due to CAFO waste management practices. These studies indicated that waste was being applied to sprayfields, then containments from waste application are leaching into groundwater. They found that groundwater flowing downgradient towards riparian buffers containing elevated nitrate concentrations, thus indicating buffers were not enough to prevent excess nitrogen species from entering nearby surface water (Karr et al., 2001; Israel et al., 2005). Karr et al. (2001)found a median of 30 mg/L of nitrate and Israel et al. found a mean of 30 mg/L of nitrate in shallow groundwater impacted by waste application. Overall, these studies conclude that high concentrations of nitrogen species in nutrient-sensitive watersheds can be linked to waste management practices at CAFOs despite best management practices like riparian buffers and current nutrient management plans. Studies like Bradford et al. (2008) and Christenson and Serre (2017) have questioned the effectiveness of current nutrient management plans to mitigate nitrogen leaching/runoff from sprayfields to water resources. Bradford et al. (2008) found that nutrient management plans may not consider waste nutrient differences within the lagoon and the nutrient uptake ability of the various nutrient compositions and sprayfields. Christenson and Serre (2017) also challenged 6 these nutrient management plans through remote sensing techniques that showed 24% of catchments in the study over a 5-year period resulted in a positive PAN balance. A positive PAN balance indicates that crops in sprayfields cannot completely assimilate PAN, potentially resulting in discharge of nitrogen-rich groundwater to nearby surface water. The goal of the nutrient management plan is to keep the PAN balance negative to ensure that labile nitrogen is used by the crops. These studies show that current waste management plans may need additional strategies to meet the goal of protecting groundwater and surface water from excess nitrogen species. 1.4. A Need for Alternative Waste Management? CAFO waste management has been under critical review and even led to costly lawsuits with large corporations, or big agriculture companies like Smithfield (Amini et al., 2017). Therefore, pressure has been applied to corporations and owner/operators of CAFOs to implement alternative waste management strategies to reduce environmental impacts. The new wastewater management strategies currently under discussion and in some cases being implemented are designed to reduce air emissions compared to open waste lagoons. Hribar (2010) summarized key operations and impacts CAFOs have in nearby areas, including air pollutants. Typical pollutants associated with CAFOs are ammonia, hydrogen sulfide, methane, fecal indicator bacteria, and particulate matter. Ammonia and hydrogen sulfide are associated with odor and human health risks like respiratory irritation, chronic lung disease, and even death. Methane is odorless and does not have associated human health effects but is a potent greenhouse gas. Agriculture has been cited as the largest source of methane in the United States (US EPA, 2018). Waste management systems can be a potentially substantial source of 7 these air emissions, especially the storage of swine waste in lagoons at CAFOs (Hribar, 2010). One alternative management strategy to address air quality concerns from swine lagoons is to cap or cover the waste lagoon to reduce emissions of nuisance odors and ammonium volatilization instead of being left uncovered and open. The capping of the lagoon may also reduce potential discharges, e.g., overflow due to flooding, directly from the lagoon to nearby surface water receptors. At the time of this thesis, there were no known studies in the literature quantifying how capping a waste lagoon might impact the nitrogen inputs to groundwater and surface water. A past study by Ducey et al. (2019) characterized differences in wastewater nitrate concentrations for an open CAFO lagoon versus one using a capping system. They found higher total Kjeldahl nitrogen (TKN) concentrations in the capped lagoon compared to the open lagoon wastewater. These results suggest that if an open wastewater lagoon is capped, then volatilization of ammonium would be severely inhibited, thereby concentrating nitrogen in wastewater and potentially reduce local nitrogen atmospheric deposition. A capped lagoon would also prevent dilution due to rainfall and therefore higher nitrogen concentrations. Land application of more concentrated wastewater from a capped lagoon may contribute to shallow groundwater and surface water degradation, especially if application rates exceed the assimilatory capacity of the sprayfield (Mallin and Cahoon, 2003). Therefore, it is important to quantify nitrogen transport from a swine CAFO using a capping system. 1.5 Organization and Study Objectives This thesis provides a unique contribution to the literature by quantifying nutrient transport in surface water and shallow groundwater resources at the site and sub watershed scale for a farm using a capping system. This thesis contains 4 chapters: introduction (1), surface water 8 analysis (2), groundwater analysis (3), and conclusion and management implications (4). Chapters 2 and 3 were formatted as standalone journal manuscripts, therefore there is redundancy between concepts covered in this chapter. Chapter 4 is written to synthesize results from chapters 2 and 3 to compare results from both chapters and to synthesize findings in the context of the larger scientific literature base. The goal of this study was to quantify nitrogen transport in water resources downgradient of a CAFO using a capped lagoon system. To achieve this goal, this thesis identified 3 specific objectives, which were to: 1) quantify nitrogen concentrations and loads in surface waters upstream and downstream of the studied CAFO (Chapter 2); 2) assess shallow groundwater nitrogen concentrations and treatment efficiency beneath and adjacent to sprayfields vs. background (Chapter 3); and 3) contextualize results in the broader literature to discuss implications from this work (Chapter 4). 9 CHAPTER 2: QUANTIFYING NITROGEN TRANSPORT TO SURFACE WATERS DOWNGRADIENT OF A SWINE CAFO IN EASTERN NORTH 2.1 Introduction Animal agriculture has been identified as the largest probable nonpoint source of nitrogen species in North Carolina rivers (Ritter, 1988). Excess concentrations of nutrients (e.g., nitrogen) can negatively affect water quality via eutrophication, harmful algal blooms, and/or fish kills (Vitousek et al., 1997; Tilman et al., 2002; Townsend et al., 2003). Water quality degradation associated with excess nitrogen can also lead to other major issues like economic damages, reduced biodiversity, and recreational impacts due to closure of beaches and other access points (Dodds et al., 2008). As the nation’s third largest swine producer, waste from animal agriculture has become a notable issue in North Carolina because of the shift from smaller traditional farms to larger, concentrated farms (Harden, 2015; USDA/NASS, 2021). Swine concentrated animal feeding operations (CAFOs) typically house at least 2,500 hogs and rely on large waste lagoons (e.g., 0.5 to 2 acres and 2.4 to 6.0 m deep) to manage wastewater generated by livestock (Mallin et al., 2015). Animal waste is generated in the hog houses and piped to open lagoons or pits where it is stored and treated anaerobically according to the CAFO’s certified animal waste management plan (Harden, 2015; Ducey et al., 2019). The waste in the lagoon is kept at a pH above 7 through liming, which promotes formation of ammonia (Mallin and Cahoon, 2003; NC General Statute § 143-215.10C). The ammonia from the waste can volatilize from the open lagoon where the waste is stored as well as from the nearby sprayfields where the waste is applied. (McCulloch et al., 1998). Increased volatilization and wet/dry deposition of ammonium as swine operations increase and expand lead to calls for more regulations regarding nitrogen emissions to prevent environmental threats to sensitive estuarine and freshwater systems in North Carolina (Walker et al., 2000). The increase of nitrogen emissions corresponded with increases in odor complaints from nearby communities due to the smell, potential health risks that lower the quality of life, and economic issues that may lower property values (Thu et al., 1997; Lu et al., 2007; Wing et al., 2008). Wastewater contains elevated plant available nitrogen (PAN), thus land application of wastewater allows recycling of nutrients by applying them to crops grown on sprayfields. Even though the amount applied is based on optimal plant uptake to prevent runoff or leaching through the soil determined by the facility’s Certified Animal Waste Management Plan, the amount permitted may not prevent nitrogen transport to water resources. This may be due to application excesses and because the industrialization of CAFOs has led to increased waste production that surpasses the assimilatory capacity of the environment (Mallin and Cahoon, 2003). Wastewater from animal feeding operations has been identified as a major source of nitrogen species in surface and ground waters because of waste management concerns (Cahoon et al., 1998; Burkholder et al. 2007; Brown et al. 2020). Past studies have found that groundwater and surface water downgradient from CAFOs and/or land application sites contained elevated nitrogen concentrations and/or masses. Stone (1998a), Harden and Spruill (2008), and Karr (2010), and Harden (2015) found higher nitrate concentrations in surface water adjacent or downgradient of swine operations than in background or upstream locations. These studies surmised that excess nitrogen was related to agricultural waste management despite swine operations applying waste at recommended rates. Swine CAFOs are also a major concern in coastal North Carolina regarding water quality as of late due to the rapid increase in swine population and size of the operations. Stone et al. (1998a) found that CAFO expansion had significant impacts on ammonia in nearby surface 11 water. Mallin and Cahoon (2003) emphasize the importance of understanding the increase and industrialization of swine CAFOs specifically because of the amount of nitrogen excreted annually compared to other animal operations. Animal feeding operations were estimated to release 124,230 tons of nitrogen annually from 1996-2001. Swine CAFOs constitute the majority nitrogen excretions from CAFOs for all types of livestock (chickens, turkey, and cattle) and contributes about 81% of the estimated total nitrogen excreted on the coastal plain in North Carolina (Mallin and Cahoon, 2003). A study by Martin et al. (2018) Landstat imagery and permit data to determine the distribution of CAFOs in North Carolina and found that 55% of the CAFOs in North Carolina (6,646) were located in the Coastal Plain. Another study by Montefiore et al. (2022) used historical Landstat 5 imagery to determine when new swine waste lagoons were built. This study found that waste lagoon construction was concentrated in the Cape Fear River watershed. To combat environmental degradation from animal waste systems, some farmers have begun to shift to new wastewater management strategies. One example of these alternative management strategies includes covering the waste lagoon with a synthetic cover to reduce odor and ammonium volatilization. A recent study (Ducey et al., 2019) found that covering waste lagoons can be effective at mitigating volatilization of ammonium. These authors reported higher TKN concentrations in the capped lagoon compared to the uncapped lagoons. Wastewater within covered lagoons contained mean concentrations of TKN (1009 ± 24 mg/L) that were more than double those observed in open lagoons (473 ± 44 mg/L), this difference could be due to the inhibition of volatilization via capping the lagoon or less dilution from direct rainfall. These results suggest that land application of wastewater from covered lagoons may pose a substantial risk to water resources, especially for nutrient-sensitive waters (Ducey et al., 2019). Therefore, 12 more information is needed to characterize nitrogen transport in surface waters downgradient of a swine CAFO using a capped lagoon. operations The goal of this study was to compare surface water quality upstream and downstream of a CAFO using a lagoon cover to evaluate sub-watershed scale nitrogen loading. At the time of this study, there were no published studies in the world quantifying physical and chemical parameters in surface water downstream from sprayfields receiving land application of wastewater from a capped lagoon. To achieve this goal, this study pursued 3 objectives, which were to: 1) quantify nitrogen concentration and exports in surface water upstream and downstream of a CAFO; 2) identify potential sources of nitrogen in surface water upstream and downstream of the CAFO; and 3) evaluate water quality impacts of a land application event using high frequency data. 2.2 Methodology 2.2.1 Study Area The study site (Fig. 2.1) is a swine CAFO in the coastal plain within the Sandhills region located in Harnett County. Annual precipitation is 115.6 cm (45.4 in) and about 70% is lost to evapotranspiration due to warmer climate and typically lower hydraulic heads associated with the coastal plain (Sun et al., 2002; O’Driscoll et al., 2010). The coastal plain is typified by soil textures ranging from clay to gravel, a shallow water table, and thickening of the surficial aquifer as it dips to the east (Winner and Coble, 1996; O’Driscoll et al., 2010). Land cover in the area is predominantly cultivated crops and hay (28%) and herbaceous upland (51%) (Appendix A) (USGS, 2022). The mean basin slope is about 7% (USGS, 2022). A stream intersects the farm and is fed by 3 unnamed tributaries, flow from groundwater seeps, and direct groundwater 13 discharge originating as flow beneath the adjacent sprayfields (Fig. 2.1). The stream discharges to the Upper Little River, which eventually feeds into the Cape Fear River. This swine CAFO has been operating since 1995 and can house up to 7,350 swine. There are two waste lagoons with an anerobic digester; the lagoons were capped in 2008 with high- density polyethylene covers. There is approximately 20.1 m (66 ft) of topographic relief from the upland area of the farm (near Sprayfield 5 in Fig. 2.1) to the lowland area of the farm (near Sprayfield 4 in Fig. 3.1). Soils at the farm are primarily Gilead loamy sand (~50%), Blaney loamy sand (~30%), and Roanoke loam (~17%) though the soils downgradient of the barns and lagoons are mainly Gilead loamy sand; the typical depth to water table for the Gilead series is 46-76 cm (18-30 inches) (USDA, 2022) (Appendix B). Table 2.1 summarizes some of the hydrologic features and soil types for each surface water location. Sprayfields in lower elevations of the farm have shallow water tables (about 30-50 cm) and finer textures that inhibit vertical infiltration. These characteristics may facilitate lateral flow within the subsurface resulting in groundwater seeps that can be a source of nutrients to surface water (Lin et al., 2005). During the current study, there were 2 – 3 groundwater seeps observed near the riparian buffer of the stream within Sprayfields 3 and 4. One of the seeps was routinely sampled because it flows perennially and eventually discharges into the stream adjacent to the farm. This seep originates as groundwater discharge at the surface between the two capped lagoons (SU on Fig. 2.1) and runs for approximately 330 m before draining to the stream (SD on Fig 2.1). 14 Figure. 2.1. Map of study area showing the surface water sampling locations. The stream intersects the farm (SW to NE). Groundwater surfaces at a perennial seep located near (SU) and flows northerly in a small channel. A small perennial creek (referred to here as East Creek) runs roughly parallel with the seep in the eastern area of the farm and drains to the stream. There is an ephemeral tributary between UP1 and UP2. A pond is located on the west side of the swine houses. Land application sites are depicted as green polygons and labeled as “Field”. 15 Table 2.1. Summary of watershed area, stream gradient, proximity to sprayfield, and the major soil series near the surface water locations. Site locations are Upstream 1 (UP 1), Upstream 2 (UP 2), Downstream (DOW), Seep-Up (SU), Seep-Down (SD), and East Creek (EC). Watershed area for SU and SD is an estimate based on topography. Site Watershed Stream Proximity to Soil Series Area (ha) Gradient Sprayfield (m) UP 1 122 0.02 120 Roanoke loam UP 2 119 0.02 70 Roanoke loam DOW 243 0.01 230 Roanoke loam SU 51 NA <1 Gilead loamy sand SD 63 0.02 70 Gilead loamy sand EC 62 0.01 <1 Roanoke loam 2.2.2 Instrumentation of Surface Water Sites Surface water quality and quantity were monitored in the stream at locations Upstream (UP) and Downstream (DOW) of the farm (Fig. 2.1). Two upstream locations (UP 1 and UP 2) were established for the purpose of this study because the UP 1 location had been observed to be dry during warmer/dry months. Water quality parameters collected from UP locations were pooled and reported as the average value. The seep was monitored near its origin adjacent to the lagoons (SU) and near its confluence with the stream (SD) as depicted in Fig. 2.1. Three Manta 35+ Sondes (Eureka Water Probes) were used during a two-week period (8/14/2020 - 8/31/2020) to monitor specific conductivity, nitrate, and ammonium. This two-week period coincided with a land application of wastewater event that occurred from August 20th - August 24th, thus water quality was quantified using high-frequency data before and during multiple spray events. The Manta Sondes recorded values for these parameters every 15 minutes and were placed at locations UP, DOW, and SD. 16 2.2.3 Sampling Frequency and Protocols Monthly water sampling took place from December 2019 to January 2021. For each sampling event, physicochemical parameters were measured in the field and water samples were collected for analysis in the Environmental Research Laboratory at East Carolina University. Due to COVID-19 lockdown protocols, there was a sampling gap from April through May 2020. A Hanna Instrument 9829 multiprobe meter was used in the field to measure pH, specific conductance (SC), temperature, dissolved oxygen (DO), oxidation reduction potential (ORP), and turbidity. Water samples for lab analysis were collected using polypropylene bottles that were rinsed in the stream or seep to prime bottles before sampling. Water samples were then transported in an iced cooler to the laboratory. The samples were immediately filtered using a vacuum filter and 1.5- and 0.7-micron Whatman glass microfiber filter paper. Each sample was then either analyzed for nitrogen concentration or frozen until these analyses could be performed. The filtrate was analyzed by a KPM Analytics SmartChem 170 or 200 discrete auto-analyzer for nitrate and ammonium and a Total Organic Carbon and Total Nitrogen analyzer (Shimadzu Scientific Instruments, Inc.) for TDN. A subset of sample filtrate was frozen and shipped overnight to the UC Davis Stable Isotope Facility for isotopic analysis of nitrate fractionation. Ratios of ?15N and ?18O in nitrate were measured at UC Davis Stable Isotope Facility using a ThermoFinnigan GasBench + PreCon trace gas concentration system interfaced to a ThermoScientific Delta V Plus isotope-ratio mass spectrometer (Bremen, Germany) (Sigman et al., 2001; Casciotti et al., 2003; Granger and Sigman, 2009). These data can provide insight into possible sources of nitrate (Showers et al., 1990; Karr et al., 2001; Silva et al., 2002; UC Davis, 2022). For example, animal or human waste typically contains elevated values of ?15N compared to other nitrate sources (e.g., fertilizers, soil 17 organic matter, atmospheric deposition) (Hübner, 1986; Silva et al., 2002). Results from UC Davis were plotted on a figure representing ranges previously established by Silva et al. (2002) for isotopic signatures for various nitrate sources. At each sampling location, a stream transect was identified based on the active stream channel to estimate stream discharge. Stream width across the transect was measured, and the stream depth was recorded approximately every 15 cm along this width to estimate mean stream depth. Mean stream velocity was measured along the transect using a GlobalWater Flow Probe 101 flow meter. If the stream velocity was too slow to engage the propeller of the flow meter, then velocity was estimated using an environmentally friendly green xanthene dye. A timer was used to record the amount of time it took the dye to travel a known distance (Dobriyal et al., 2016). The cross-sectional area of the active stream channel was calculated by multiplying the stream width by the mean stream depth. This area was then multiplied by the mean stream velocity measured to calculate discharge. 2.2.4 Statistical Analysis Nutrient concentrations, watershed exports, and physicochemical parameters were compared via statistical analysis and visually represented using the R statistical framework (R Core Team, 2019). Comparisons were made between UP vs. DOW, SU vs. SD, and seep water quality parameters relative to UP and DOW. Data normality was tested using a Shapiro-Wilks test in R (Appendix C) (Shapiro and Wilk, 1965). If the data were not found to follow a normal distribution, the data were transformed and tested again to determine if assumptions of normality were met. Only some data sets met normality assumptions post transformations, so all hypothesis tests were performed using non-parametric methods for consistency. Hypotheses were tested using the Kruskal-Wallis method to detect if any significant differences (p < 0.05) exist between 18 categorical variables. If a significant difference was found, then a post-hoc pairwise Wilcoxon rank sum test with the Bonferroni p-adjustment factor was run to determine statistical significance between comparison groups (Wilcoxon, 1945; Kruskal and Wallis, 1952). Nitrogen loads were calculated by multiplying monthly discharge and nutrient concentration data to compare differences in mass flux between upstream and downstream of the farm. Additionally, nitrogen loads were calculated at the SU and SD sampling locations to estimate in-seep processing of nitrogen and quantify seep contributions to the DOW location. 2.3 Results and Discussion 2.3.1 Nitrogen Concentrations TDN concentrations in surface water suggested that the farm substantially affected surface water quality (Fig. 2.2). The studied seep had the highest TDN concentrations of all surface water sampling locations. The median concentration of TDN in the seep was 54.41 mg/L and 37.29 mg/L for SU and SD, respectively, which was found to be a statistically significant difference (p= 0.04) (Table 2.2). The SU location had higher concentrations of TDN than SD during 92% of sampling events (12 out of 13 events) (Appendix D). While there was a 32% reduction in median TDN concentration between SU and SD in the seep, TDN concentrations at SD were substantially greater than those in the stream at DOW and UP stream TDN concentrations (p < 0.01). Samples collected downstream of the farm at DOW contained a median TDN concentration of 6.65 mg/L, which was about 14 times higher than the influent surface water sampled at UP (0.48 mg/L; p < 0.01) (Fig. 2.2). TDN concentrations at SD were greater than DOW during 85% of the times sampled (11 out of 13). Samples from downstream of 19 the farm at DOW contained TDN concentrations that exceeded those upstream of the farm at UP 92% of the times sampled (12 out of 13). Figure. 2.2 Ammonium, nitrate, and total dissolved nitrogen (TDN) concentrations collected for each surface water location: Upstream (UP), Downstream (DO), Seep-Up (SU), and Seep-Down (SD). 20 Figure. 2.3. Time series showing ammonium (NH +4 ), nitrate (NO -3 ), dissolved organic nitrogen (DON), and total dissolved nitrogen (TDN) for Upstream (UP), Downstream (DOW), Seep-Up (SU), and Seep- Down (SD). 21 Plant available forms of nitrogen (primarily nitrate) were the most prevalent N species downgradient of land application sites. The most dominant species of nitrogen observed in surface water locations downgradient of the CAFO was nitrate, which made up 18%, >99%, >99%, and 93% of the median TDN values of UP, DOW, SU, and SD, respectively (Fig 2.2). Ammonium (NH +4 ) made up 11%, <1%, <1%, and <1% of the median TDN values of UP, DOW, SU, and SD, respectively. The remaining TDN in the UP (71%) and SD (7%) is made up of dissolved organic nitrogen, which was estimated based on the differential between TDN and dissolved inorganic nitrogen (DIN) (nitrate and ammonium). Surface waters downgradient from sprayfields were comprised of mostly nitrate throughout the study period (Fig. 2.3). Seasonally, DOW, SU, and SDhad the highest TDN concentrations during the growing seasons (e.g., March- September), which is when CAFO operators can irrigate sprayfields with wastewater (Fig. 2.3). Overall, the current study found TDN concentrations ranged from 0.33 – 7.67 mg/L upstream of the farm (UP), 2.28 - 40.69 mg/L downstream of the farm (DOW), and 14.86 - 60.91 mg/L in the seep on the farm (SU and SD). The TDN concentration of 7.67 mg/L on February 17, 2022 at UP was likely an outlier and removing it alters the range to 0.33 - 3.14 mg/L. The elevated TDN concentrations observed in the seep and DOW surface water locations relative to UP of the farm suggest that the farm is a significant source of TDN to local water resources, and possibly to the Cape Fear River. Past studies suggest that CAFOs can be a significant source of nutrients to nearby water resources (Evans et al., 1984; Stone et al., 1995; Harden, 2015; Mallin et al., 2015). Collectively, these studies found TDN concentrations downgradient of CAFOs ranging from 3.6 – 38.5 mg/L. Mallin et al. (2015) found that TN ranged from 3.8 – 13.4 mg/L in surface water locations within 3 km of a CAFO in the Cape Fear River Basin. Harden (2015) reported higher values of TN concentrations that ranged from 3.6 – 38.5 mg/L in stream 22 sampling sites located downstream of nearby CAFOs. Harden (2015) likely found a higher maximum than Mallin et al. (2015) due to the proximity to the farm. Surface waters adjacent to and downgradient of the sprayfields contained TDN concentrations that may pose an environmental health risk. More specifically, studies have suggested that TN concentrations above 0.8 - 1.5 mg/L may facilitate phytoplankton growth (Dodds et al., 1998; Xu et al., 2015; Zeng et al., 2016). The median TDN concentrations reported here for SU, SD, and DOW had median TDN concentrations that exceeded 0.8-1.5 mg/L during all sampling events. Seep and DOW locations contained TDN values high enough to initiate eutrophication (Fig. 2.2), which may result in harmful algae blooms and fish kills. These locations also exceeded US EPA TN reference conditions (conditions without human impact) for Ecoregion IX (range of the 25th percentile: 0.07 – 1.0 mg/L) while UP exceeded 1.5 mg/L only 19% of the time. These reference conditions are important because they can be used to compare results to possible conditions estimated by the EPA to determine if there are human or environmental impacts. The differences in maximum TDN values in the current study versus previous studies may be due to the proximity of the sampling sites to the farms. As the distance from a CAFO increases downstream, there is increased opportunity for instream processing to remove nutrients or for dilution to reduce concentrations. Instream processing may facilitate nitrogen removal through plant uptake, denitrification, and immobilization (Saunders and Kalff, 2001; Cheng and Basu, 2017). In the current study, a reduction of TDN was observed between SU and SD and between SD and DOW. Instream processing and dilution are both viable reduction mechanisms. Instream processes can reduce nitrogen concentrations through plant uptake and can occur in the hyporheic zone. A study by Storey et al. (2004) found that the hyporheic zone of a stream (zone 23 immediately beneath the stream bottom) can be an efficient nitrate sink. Dilution, plant uptake, immobilization, and denitrification can all reduce TDN concentrations in surface water. The median concentration of nitrate in the SU and SD sampling locations exceeded the 10 mg/L nitrate-nitrogen drinking water standard regulated by the US EPA (2002). Although these waters are not designated as water supply, these data show that surface water located near sprayfields can contain elevated nitrate concentrations that endanger public health (Sadeq et al. 2008). Nitrate concentrations in the SU and SD sampling locations exceeded this standard 100% of the time (n=13). The DOW location contained a median nitrate concentration (6.16 mg/L) that did not exceed the drinking water standard, however nitrate concentrations were observed to exceed the standard 31% of the time (4 out of 13). The decline in the median nitrate concentration between the SD and the DOW suggest that instream processes further reduced nitrate concentrations, nitrate transport from the seep to the stream may not be substantial, dilution may be occurring, or some combination thereof (Royer et al., 2004; Udy et al., 2006). In addition to environmental and public health issues, elevated nitrate in surface waters requires more energy, and by extension increases the cost, to treat potable water to meet the Safe Drinking Water Act nitrate-nitrogen standard. A study by Twomey et al. (2010) found that increase in corn cultivation in farming to create biofuels could lead to a 2100% increase in the energy needed to remove the additional nitrate that will be put in the environment. High-frequency DIN concentration data collected before and during a spray event showed increases in DOW and SD compared to UP of the CAFO (Fig. 2.4). Table 2.2 shows the days spraying occurred with the amount of waste applied, how long waste was applied, and the sprayfield location. The closest sprayfields to the stream are 1, 2, and 4, which were mostly sprayed within the first few days (August 20-22). The median DIN values during this two-week 24 period prior to the spray event (August 14th-20th) were 1.0, 3.4 , and 25.3 mg/L for the UP, DOW, and SD locations, respectively. DIN data measured by the sondes were within the 2 mg/L or 5% range of the samples analyzed in the lab at this time. The waste stored in the lagoon sprayed during this time had a TDN concentration of 985.3 mg/L (Chapter 3). During and after the spray events, there were increased median DIN values of 6.8 mg/L and 41.9 mg/L in the DOW and SD locations, respectively, while the UP remained at or near 1 mg/L. DOW of the CAFO showed two major spikes on August 24th and August 26th at 137.9 mg/L and 82.6 mg/L, respectively. These spikes observed in the high frequency data are much larger than the previously mentioned range from the intermittent (i.e., monthly) samples in this study (0.33 – 75.54 mg/L) or other studies (3.6 – 38.5 mg/L) (Evans et al., 1984; Stone et al., 1995; Harden, 2015; Mallin et al., 2015). The spikes observed in DOW were short lived (~1 day duration) and would be difficult to capture via intermittent grab sampling, which may be why the spikes observed at DOW with the in-situ DIN sensor were high in comparison. The seep had cycles of low TDN corresponding with high solar radiation and algal uptake and higher TDN during the day when solar radiation and algal uptake are more likely occurring (Huang et al., 2022). These responses are likely not due to precipitation during this time (Appendix E). Diurnal responses occurred whether or not there was precipitation. Diurnal responses within the seep also indicate time of grab sample during the day may impact TDN concentration results. 25 Figure. 2.4. Time series data dissolved inorganic nitrogen (DIN) concentrations in the UP, DOW, and SD locations every 15 minutes during two weeks before and after a spray. The spray event is shown in gray and occurred daily from August 20th to August 24th. There were no major rain events during this time (Appendix E). 26 Table 2.2. Spray field irrigation by date, location, time, and total volume of land applied wastewater (Sousan et al. 2021). Date Spray Field Total Time Total Volume (Minutes) (Liters) August 20 1 295 307,436 August 20 2 90 61,324 August 21 1 445 409,642 August 21 2 90 61,324 August 21 4 330 224,853 August 21 5 300 204,412 August 22 1 150 102,206 August 22 2 120 81,765 August 22 3 450 306,618 August 22 4 330 224,853 August 22 5 570 388,383 August 23 2 120 81,765 August 23 3 450 306,618 August 23 5 420 289,448 August 24 5 150 105,477 2.3.2 Sources and Pathways of TDN The results for samples sent for isotopic analysis are shown in Fig. 2.5 relative to typical ranges of ?15N and ?18O for different sources of nitrate (Kendall and McDonnell, 1998; Silva et al., 2002). The isotopic data suggest that the most likely source of nitrate in all the sampling locations was animal or human waste. Human waste is not a likely source since residential development in the study area is sparse (only 8 houses). While these houses are likely served by septic systems, the closest potential septic system is about 145 m away, so this is unlikely to be a substantial source. For instance, Humphrey et al. (2016) found a mean TDN concentration of 1.25 mg/L downgradient 35 m from a septic system, though the conditions in that study may not be representative of those in this study. Upland areas of the basin are predominantly forested with little residential development, further supporting that human waste sources are unlikely. Therefore, the most likely source of nitrate was animal waste. Swine waste was the most likely 27 source in SU, SD, and DOW based on isotopic fractionation, land cover, and recently discussed nitrogen concentrations. Interestingly, the isotopic data also suggested that surface water upstream of the farm at location UP was also most likely related to human or animal waste. It is therefore possible that dominant nitrate sources in upstream forested areas could be due to wildlife (Cerling et al., 2004). Wildlife (e.g., deer, rabbits, snakes, alligators, birds, etc.) was commonly observed at the field site by researchers, thus wildlife was present and their defecate could be a potential source of nitrate after microbial processes convert organic nitrogen to nitrate. Figure. 2.5. Stable isotope data for the Upstream (UP), Downstream (DOW), Seep-Up (SU), and Seep- Down (SD) surface water locations. Source fields of nitrate were derived from previous studies (Kendall and McDonnell, 1998; Silva et al., 2002). NP=nitrogen precipitation, SNF= synthetic nitrate fertilizer, AF= ammonium fertilizer, SN=synthetic nitrogen, AHW=animal or human waste. 28 Conductivity has also been used as a tracer for sources of waste (Birch et al., 2016). The highest conductivity values were observed in the seep with median conductivity values of 721 µS/cm at SU and 659 µS/cm at SD (Table 2.3). UP had the lowest conductivity values with a median of 29 µS/cm and DOW had a median of 144 µS/cm (Table 2.3). Median values of conductivity differed significantly (p < 0.01) between all locations except between the SU and SD (p = 0.46). The median values of conductivity observed in the stream show similarities to other studies findings. Harden (2015) reported similar trends where conductivity downstream of CAFOs was elevated compared to downstream watersheds without CAFOs present. Harden (2015) reported a median value of 98 ?S/cm in background locations and 138 ?S/cm downstream from swine CAFOs. Other studies, such as Katz et al. (2009) and Birch et al. (2016), found wastewater and water influenced by wastewater had elevated conductivity values. Katz et al. (2009) reported SC values in waste of 524 ?S/cm and 613 ?S/cm in waste that was applied to sprayfields nearby. The median conductivity values observed at the seep in this study are either similar or elevated compared to those measured in other studies, further suggesting wastewater impacts. Birch et al. (2016) found wastewater had the highest specific conductance values (approximately 700 ?S/cm), which is also similar to conductivity in the seep but much higher than the conductivity values observed in UP and DOW. SC values observed in the seep and DOW locations suggest that wastewater is a likely source of nutrients, which is congruent with isotopic fractionation data that was previously discussed. 29 Table 2.3. Physical and chemical parameters collected at the farm through monthly sample events. Median (range) of discharge (Q), normalized discharge (Q), pH, temperature, specific conductivity (SC), dissolved oxygen (DO), turbidity (Turb.), and oxidation reduction potential (ORP). Site Q Q pH Temp. SC DO Turb. ORP (L/sec) (L/ha/sec) (°C) (µS/cm) (mg/L) (FNU) (mV) UP 16.8 0.14 3.6 11.4 29 7.70 9.3 276.65 (0.6-67.7) (0.01-0.41) (2.78-5.05) (8.5-23.4) (0-48) (4.98-11.10) (0-30.5) (122-496) DOW 35.7 0.15 5.2 13.3 144 8.67 12.2 220.6 (2.3-141.6) (0.01-0.58) (4.65-6.59) (9.2-24.7) (86-805) (5.70-12.5) (4.6- (101-614) 28.5) SU 4.0 0.08 4.17 16.6 720.5 8.60 0.9 229.3 (1.1-9.1) (0.02-0.18) (3.77-5.37) (9.5-26.6) (356-985) (5.19-10.95) (0-15) (119- 447.5) SD 2.5 0.04 4.67 17.4 658.5 9.28 4.70 229 (0.6-17.0) (0.01-0.3) (4.04-7.57) (8.2-28.9) (191-793) (4.83-13.15) (0-32.8) (73.8-439) During a monthly sampling event on August 31st, 2020, a longitudinal conductivity survey of the stream was conducted, and 24 SC measurements were collected between the UP and DOW sampling locations (Fig. 2.6) (Appendix F). This analysis shows that conductivity values increased downstream of the confluence between the stream and the seep. SC values ranged from 20 – 40 ?S/cm, which was similar to UP values (0-48 ?S/cm), until downstream of the confluence of the seep (Table 2.3). The SC showed additional spikes at the intersection of the stream and the East Creek as well. SC values spiked from 14 ?S/cm to 917 ?S/cm and remained elevated relative to UP until another spike of 587 ?S/cm at the discharge location of the East Creek. This spike in SC at the seep and East Creek could indicate that groundwater seeps and/or other drainage features on the farm are potential pathways for wastewater-derived nutrients and ions that increase the capability of water to pass electrical flow (Del Rosario et al., 2014; Humphrey et al., 2014; O’Driscoll et al., 2019). 30 Figure. 2.6. Choropleth map of longitudinal conductivity survey conducted on August 31, 2022. Greater conductivity values correspond to larger circle size (Appendix F). 2.3.3 Surface Water Discharge and Exports The median discharge of water observed at each site was 16.8 L/sec, 35.7 L/sec, 4.0 L/sec, and 2.5 L/ sec in the UP, DOW, SU, and SD, respectively (Table 2.3). SU and SD were not significantly different (p = 1), but both were significantly different from UP (p < 0.01) and DOW (p < 0.01). Watershed area (Table 2.1) for the seep was estimated conservatively based on topography (Appendix G). When discharge was normalized by the estimated watershed area for the seep, there was no significant difference between SU (0.08 L/ha/sec) and SD (0.04 L/ha/sec) (Table 2.3). UP and DOW were significantly different from one another (p < 0.01), but this may be due to watershed area differences. Table 2.3 shows watershed differences and median 31 normalized flow of 0.14 L/ha/sec and 0.15 L/ha/sec for UP and DOW, respectively. When normalized by watershed areas, UP and DOW were not significantly different (p = 0.41). This indicates that the UP watershed is contributing similar discharges as the downstream watershed that contains the farm. Temporal differences were observed in the stream for normalized discharge in UP and DOW (Fig 2.7). Discharges tended to be lower during warmer months (March-August), whereas precipitation tended to be greater during these months (Fig 2.8.) Seasonal trends observed in the stream are most likely due to high evapotranspiration in the summer that draw down water tables resulting in lower discharge based on studies like Harden and Spruill (2008) that show streams are commonly fed by groundwater in the coastal plain (Spruill et al., 1998; O’Driscoll, 2012). Evapotranspiration in the summer outweighs the precipitation causing lower hydraulic heads, baseflows, and ultimately lower stream discharge since the water table is being drawn down (Harden and Spruill, 2008; O’Driscoll, 2012). The NC coastal plain loses about 70% of annual precipitation to evapotranspiration according to Sun et al. (2002) and O’Driscoll (2012). 32 Figure. 2.7. Normalized discharge for UP and DOW locations showing temporal trends with higher discharge in colder months (November-February). 33 Figure. 2.8. Monthly precipitation data from December 2019 to January 2021(A) and monthly temperature of surface water locations (B) showing warmer temperatures trending with high precipitation. 34 Mass loads of TDN based on baseflow yielded similar results as concentration data suggesting that the farm was a source of nutrients to receiving surface waters (Fig. 2.9). The median TDN export at UP was 316.7 kg/yr, which was significantly lower than all other locations (p < 0.01). The median mass export of TDN at DOW was 7818.4 kg/yr, which was greater than at SU and SD but only significantly so from SD (p < 0.01). SU and SD contained median exports of TDN that were 4794.0 and 1804.7 kg/yr, respectively (Fig. 2.9 A), and this difference was not significant (p = 0.63), though there was a 62% decrease in median TDN exports between the SU and the SD location. Based on median values of SD exports and DOW exports, the seep could contribute up to 25% of the DOW exports assuming limited instream processing between these sites. The relative percentage between these sites was highly variable ranging from 6% to 561%; however, it was estimated that for most sampling events (8 out of 13), SD contributed 6% - 33% of TDN export from DOW. These exports are based on baseflow, and it is possible that storm events would result in greater TDN exports. One potential reason for differences in exports could be the watershed area which can be used to normalize exports. Watershed area for the seep was estimated based on topography since the exact watershed area was unknown (Appendix G). When normalized by watershed area, there was 70% decrease between SU median TDN export (127.1 kg/yr/ha) and SD median TDN export (38.5 kg/yr/ha) but this was not significantly different (p = 0.24) (Table 2.4). Mass export of TDN was also reported as area-normalized exports to correct for differences between the UP and DOW sub-watershed area. DOW of the farm had a median area-normalized TDN export (32.1 kg/ha/yr) more than 11 times greater than UP (2.8 kg/ha/yr) (Fig. 2.10 B), and this difference was statistically significant (p < 0.01). Based on SD median area-normalized TDN 35 export (38.5 kg/yr/ha) and DOW median area-normalized TDN export (32.1 kg/ha/yr), there is likely treatment occurring between the seep location and downstream. Assuming the UP location represents background conditions, then the difference between DOW and UP TDN exports represent the TDN inputs from land application of wastewater. Therefore, the farm exported 7,501.7 kg/yr or 29.5 kg/ha/yr at the DOW location, which represented >90% of the total TDN export from the studied sub-watershed. Based on this simple mass budget, the farm is contributing 11 times the amount of TDN exports than upstream. TDN exports downstream of the farm were similar to other agricultural exports based on literature. Other studies in agricultural watersheds found TDN exports of 9.4-35 kg/yr/ha (Deal et al., 1986; Jordan et al., 1997; Lowrance et al., 1984; Peterjohn and Correll, 1984). Another study from Schilling and Zhang (2004) found similar annual mean nitrate exports from an agricultural watershed in Iowa. The study found a mean annual nitrate export of 26.1 kg/ha which was similar to exports estimated from this farm (29.5 kg/ha/yr). Table 2.4. Summary of median and range (in parenthesis) concentrations for chloride (Cl-), ammonium (NH +4 ), nitrate (NO -3 ), total dissolved nitrogen (TDN), TDN load, and TDN loads (normalized). Normalized TDN loads for the seep are based on a conservative estimate of watershed area from topography. Cl- NH +4 NO -3 TDN TDN Load TDN Load Site (mg/L) (mg/L) (mg/L) (mg/L) (kg/yr) (kg/yr/ha) 4.06 0.05 0.09 0.45 316.7 2.8 UP (1.97-14.61) (0.01-7.59) (0.02-0.44) (0.1-7.67) (25.0-4455) (0.2-37.4) 15.24 0.13 6.59 6.65 7818.4 32.1 DOW (9.41-47.22) (0.05-13.71) (2.23-30.05) (2.28-40.69) (1674.6-25534.2) (6.9-139.8) 55.65 0.645 53.99 54.77 6482.1 127.1 SU (26.17-75.61) (0.18-1.02) (21.75-74.78) (22.11-75.54) (1210.3-17517.2) (23.7-341.5) 43.47 0.33 33.71 36.38 2426.0 38.5 SD (13.49-58.16) (0.06-3.96) (9.78-59.91) (13.66-60.06) (298.9-10048.7) (4.7-159.5) 36 Figure. 2.9. Plot A shows TDN exports (kg yr-1) for all surface water locations. Plot B shows TDN exports (kg ha-1 yr-1) normalized based on drainage basin size. 37 2.4 Conclusion The goal of this study was to evaluate surface water quality impacts downstream from a capped lagoon system. Results indicate negative water quality impacts downstream based on nitrogen concentrations, exports, potential sources, and water quality response to land application. Median TDN concentrations were about 13x higher downstream (DOW median: 6.65 mg/L) than upstream (UP median: 0.48 mg/L). Normalized median TDN exports were also 11x higher downstream (DOW median: 32.1 kg/ha/yr) than upstream (UP median: 2.8 kg/ha/yr) of the farm. Downstream TDN exports were similar to other studies in agricultural watersheds that found TDN exports of 9.4-35 kg/yr/ha (Deal et al., 1986; Jordan et al., 1997; Lowrance et al., 1984; Peterjohn and Correll, 1984). The samples from the seep emerging adjacent to the waste lagoons (SU and SD) had elevated TDN concentrations and exports compared to the upstream background site (UP). SD had a lower normalized discharge than SU, indicating the seep acts as a losing stream. Because of this, there were also reductions within the exports between SU and SD. Median TDN concentrations at the seep sites were 54.77 and 36.38 mg/L and median TDN exports were 4793.97 kg/yr and 1804.70 kg/yr at SU and SD, respectively. Despite this reduction, SD still contributes elevated TDN concentrations and exports relative to UP. The seep is therefore estimated to have contributed a median of 25% of TDN exports to the outflows at DOW. The amount varied over the course of the study, ranging from 6% to 561%. Based on the 25% median contribution, the seep may not always be the most significant source of nutrients to the stream. The seep contribution to the stream may be variable due to waste application during spray events. The data collected for this study was also based off baseflow, inputs like storm events would likely lead to greater nitrogen loading than what was captured with this study. 38 Isotopic fractionation data and land cover indicate the source of nitrate found in the surface water is likely animal waste. The isotopic data alone indicated animal or human waste, however human waste is unlikely based on the land use in the area. Electrical conductivity measurements have also been used as an indicator for waste in past studies (Birch et al., 2016) and was found to be elevated within the surface water locations sampled on the farm. The median conductivity at DOW was 144 µS/cm and was elevated compared to UP, which had a median of 29 µS/cm. The highest conductivity values were observed in the seep with median conductivity values of 721 µS/cm at SU and 659 µS/cm at SD. A longitudinal survey of the SC within the stream showed increased SC between the seep and DOW, indicating the seep as potential conduit of nutrients to the nearby stream. The excess nutrients and waste are likely from land application on the sprayfields based on data and trends. High-frequency data collected using an in-situ DIN probe during a spray event showed extreme responses in nitrogen in the seep and at DOW. These data were collected over two weeks and indicated that surface waters downgradient of sprayfields, i.e., where the waste was applied, were substantially influenced by land application events. The data from the DOW site had two spikes after the spray event that reached 137.9 and 82.6 mg/L. The seep showed an increase in TDN from a median of 25.3 mg/L pre spray event to a median of 41.9 mg/L. This farm contained several seeps that were not monitored for this study, thus it is possible that other seeps may also be substantial sources of nitrogen to the studied stream and other water resources in the Cape Fear River Basin. Furthermore, groundwater beneath sprayfields may also contain elevated nitrogen concentrations and act as another potential pathway of nitrogen to surface waters, which is explored further in Chapter 4. 39 Elevated nitrogen concentrations from agricultural farms are a major concern in coastal North Carolina. NC DEQ (2018) has identified nonpoint sources, like agriculture, being the largest contributor of limiting nutrients. The current study found that CAFOs have potential to be significant contributors of nutrients to surface waters, thus future work should focus on alternative management approaches to improve nutrient management and mitigate water quality degradation. Prior treatment to the lagoon waste before land application of best management practices could help reduce water quality impacts. Implications of this work suggest one method to reduce nitrogen exports could be to implement best management practices designed to facilitate nutrient removal processes (e.g., denitrification) through use of denitrifying bioreactors or restore, create, or modify wetlands. Subsurface bioreactors have been found to reduce total nitrogen concentrations by 72% and nitrate concentrations by 99% according to a study by Husk et al. (2017). Instream bioreactors are similar to subsurface bioreactors and have also had success at reducing nitrogen concentrations. Robertson and Merkley (2009) found nitrate reductions from 4.8 mg/L to 1.04 mg/L using an instream bioreactor and wood chips to promote denitrification. Strategies applied to this specific farm would likely be effective at reducing nitrogen inputs at other CAFOs. 40 CHAPTER 3: ANIMAL WASTEWATER MANAGEMENT IMPLICATIONS BASED ON GROUNDWATER NITROGEN CONCENTRATION AND EXPORT IMPACTS AT A SWINE CAFO IN EASTERN NORTH CAROLINA 3.1 Introduction Increased availability of plant nutrients (e.g., nitrogen) is a major issue in the waters of coastal areas. As a limiting nutrient, nitrogen can cause negative environmental impacts like eutrophication, algal blooms, and fish kills (Vitousek et al. 1997; Tilman et al. 2002; Townsend et al. 2003). Human health impacts like methemoglobinemia (or “blue baby syndrome”) and cancer can also be caused by increased nitrogen consumption (Ward et al., 2005). Eutrophication and its effects not only degrade water quality but can also contribute to economic damages, reduced biodiversity, and recreational impacts (Dodds et al., 2008). Agriculture has been identified as the largest probable nonpoint source of nitrogen in North Carolina (Ritter, 1988; Stone et al., 2003). Both row crop and animal agriculture can be significant sources of nitrogen species, especially as demand for agricultural products grow (Harden, 2015). Smaller, traditional farms were commercialized into concentrated animal feeding operations (CAFOs) to meet growing populations and increased demand for animal products (Mallin and Cahoon, 2003). North Carolina defines a swine operation as a “CAFO” if it houses at least 250 swine (NC G.S. 143-215.10B). While commercialization of livestock production has provided the United States with access to affordable animal products, one of the most challenging environmental externalities is animal waste management. North Carolina is the nation’s third largest swine producer (USDA/NASS, 2021), thus swine waste management is an important and contentious environmental challenge (Harden, 2015). A typical swine CAFO consists of four main components, including hog houses to store the animals, lagoons where the waste is stored, sprayfields where the waste is applied, and setback distances between sprayfields and water sources. Most swine CAFOs house at least 2,500 hogs and rely on large waste lagoons (e.g., 0.5 to 2 acres and 2.4 to 6.0 m deep) to manage wastewater generated by livestock (Mallin et al., 2015). Typically, animal waste is generated in the hog houses and then transported to waste lagoons or pits where it is stored and treated anaerobically according to the CAFO’s certified animal waste management plan (Ducey et al., 2019; Harden et al., 2015). The waste in the lagoon is kept at a pH above 7 through liming, which promotes formation of ammonia (Mallin and Cahoon, 2003; NC General Statute § 143- 215.10C). Setback distances are established between sprayfields and water sources to mitigate water contamination (100 ft) (US Environmental Protection Agency, 2012). A permit may be required if the CAFO exceeds 2,500 swine that weigh more than 55 pounds (U.S. Code § 1251). Animal wastewater can negatively affect water resources beneath and downgradient of CAFOs because swine waste contains elevated concentrations of ammonium, which is a plant- available nitrogen. Therefore, land application of wastewater on sprayfields provides an opportunity for recycling of waste-derived nitrogen by crops via plant uptake. If wastewater is applied at a volume that exceeds the nutritive needs of the plant or applied during inclement weather, nutrients may leach into groundwater and eventually discharge into nearby surface water (Evans et al., 1984; Harden, 2015). Thus, swine wastewater can be a significant source of nitrogen in groundwater and surface water downgradient from CAFOs. These elevated concentrations and/or masses of plant-available nutrients pose environmental health risks, especially in nutrient-sensitive waters (Cahoon et al., 1998; Burkholder et al. 2007; Brown et al. 2020). 42 Past studies have found that groundwater downgradient from CAFOs and/or sprayfields contained elevated nitrogen concentrations and/or masses (Table 3.1). Stone et al. (1998a; 1998b), Karr et al. (2001), and Israel et al. (2005) found higher nitrate concentrations in groundwater beneath and/or downgradient of sprayfields when compared to background or upstream locations. These studies surmised that excess nitrogen was related to animal waste management despite swine operations applying waste at recommended rates and found nitrogen concentrations ranging from 0.1 mg/L - 54.2 mg/L (Table 3.1). Mallin and Cahoon (2003) estimated that CAFOs excreted 101,000 metric tons of nitrogen annually from 2000-2001, which accounted for approximately 81% of total nitrogen exports (124,230 metric tons) from all animal feeding operations. Stone et al. (1998b) discusses how location can impact nitrogen transformations and elevated concentrations beneath sprayfields in groundwater as well. Spatial patterns may exist based on characteristics like grain size or soil texture. Soil texture can impact nitrate leaching because coarser grains (e.g., sandy soils) allow more nitrate to leach into groundwater verses finer textures or clay soils since coarser textures do not have the same water retention capacity (Donner et al., 2004). Climate and weather also vary spatiotemporally, and high precipitation can increase nitrate leaching if the precipitation exceeds evapotranspiration (Steenvoorden et al., 1986). For example, coastal plain provinces typically contain greater sand content in shallow subsoils, which increases the potential for nitrate leaching. Nitrate leachability could become exacerbated if application occurred before, during, or immediately after periods of increased precipitation. Land application volumes and rates of waste are based on a facility’s certified animal waste management plan, which includes the realistic yield expectation of the crop, the crop’s ability to uptake nutrients, the soil series below the sprayfield, concentration of plant available 43 nutrients (PAN) in wastewater, and timing of waste application (i.e., fair-weather conditions). The strategies developed for a management plan should mitigate nutrient leaching or runoff from sprayfields to groundwater and surface water. However, the amount and rate of land application may not prevent nitrogen impacts on water quality because the industrialization of CAFOs has led to increased waste production that may surpass the assimilatory capacity of the environment (Mallin and Cahoon, 2003). To combat environmental degradation from animal waste systems, some farmers have begun to shift to new wastewater management strategies. One example of these alternative management strategies includes covering the waste lagoon to minimize gas exchange between the lagoon and atmosphere, which can reduce odors and volatilization of ammonium (Sousan et al., 2021). A recent study (Ducey et al., 2019) found that covering waste lagoons may concentrate total Kjeldahl nitrogen (TKN) relative to an uncovered lagoon. Ducey et al. (2019) reported higher mean TKN concentrations (1009 mg/L ± 24 mg/L) in the capped lagoon compared to an uncapped lagoon (473 mg/L ± 44 mg/L). Covered lagoons contained more than double mean TKN concentrations than what was observed in open lagoons, which may be due to the inhibition of volatilization. These results suggest that covering lagoons could concentrate ammonium in lagoon wastewater, which affects the volume of wastewater that can be land applied. If land application volumes/rates do not account for these differences, then water quality downgradient from sprayfields can be substantially degraded, especially those within nutrient-sensitive watersheds (Harden, 2015). At the time of this study, no other studies were conducted within or downgradient of sprayfields receiving land application of wastewater from a covered lagoon. Therefore, this study is important to further the understanding of nitrogen transport in groundwater downgradient of a swine CAFO using a capped lagoon. 44 The goal of this study was to characterize nitrogen transport in shallow groundwater beneath sprayfields receiving wastewater from a capped lagoon system. To achieve this goal, 3 objectives were pursued: 1) quantify nitrogen concentrations in groundwater beneath sprayfields and lagoon wastewater; 2) estimate concentration reductions of nitrogen in groundwater beneath and adjacent to sprayfields relative to the lagoon; and 3) estimate mass reductions of total dissolved nitrogen (TDN) using a two-component mixing model. 45 Table 3.1. Summary of literature and findings for nitrogen and specific conductivity. Source Findings Physiographic Setting Groundwater within sprayfields was impacted by swine waste application. Nitrate Stone et al. (1998a) concentrations ranged from 0.15 to 12.1 mg/L Eastern Coastal Plain (US) Groundwater with elevated nitrate due to waste application was discharging to surface water. Karr et al. (2001) Median nitrate concentrations ranged from 0.3 Mid-Atlantic Coastal Plain (US) to 30.0 mg/L. Groundwater in shallow groundwater below sprayfields had elevated nitrate after new Israel et al. (2005) regulations. Nitrate concentrations ranged from Eastern Coastal Plain (US) 6 to 35 mg/L. Elevated nitrate levels in groundwater that exceeded safe drinking water levels due to high concentration of animal waste application and Stone et al. (1998b) Eastern Coastal Plain (US) excess waste applied. Nitrate averages ranging from 0.1-54.2 mg/L. Higher mean TKN concentrations (1009 mg/L) in capped lagoons compared to uncapped lagoons (473 mg/L). These differences could Covered Lagoon: Eastern NC Ducey et al. (2019) be due to the inhibition of ammonium Uncovered Lagoon: not listed volatilization from the capped lagoon Elevated specific conductivity values in Michalopoulos et al. (2014) groundwater beneath CAFO sprayfields Crete, Greece ranging from 814-1965 µS/cm. Typical background nitrate levels in Mueller and Helsel (1996) groundwater are below 2 mg/L based on 60 Throughout the United States large watershed and aquifer systems. Nitrate concentrations beneath CAFO Spruill et al. (2002) sprayfields ranged from 10- 35 mg/L with a Eastern Coastal Plain (US) median of 26 mg/L. Swine waste lagoon total nitrogen Miner et al. (2003) concentrations ranging from 327-756 mg/L. Eastern Coastal Plain (US) Lagoon waste TKN ranging from 1590-1727 VanderZaag et al. (2010) mg/L in an open lagoon and 1150-1783 mg/L in a closed lagoon. Nova Scotia, Canada 46 3.2 Methodology 3.2.1 Study Area The study site (Fig. 3.1) is a swine CAFO located in the coastal plain and Sandhills region of Harnett County. Annual precipitation is 115.6 cm (45.4 in) and about 70% is lost to evapotranspiration due to warmer climate and typically shallow water tables associated with land surface close to sea level (Sun et al., 2002; O’Driscoll et al., 2010). The coastal plain is typified by shallow water tables with surficial aquifers dipping and thickening to the east and soil textures ranging from clay to gravel (Winner and Coble, 1996; O’Driscoll et al., 2010). Most of the land cover is cultivated crops (28% hay) and herbaceous upland (51%) (Appendix A). The mean basin slope is about 7% (US GS, 2022). The farm containing the CAFO in this study is intersected by a tributary to the Upper Little River, which receives discharge from 3 unnamed tributaries, diffuse flow from seeps, and groundwater beneath adjacent sprayfields. After discharging into the Upper Little River, surface water eventually feeds into the Cape Fear River (approx. 14 km) downstream from the farm), which is one of North Carolina’s largest streams that ultimately outlets into the Atlantic Ocean. The studied swine CAFO has been in operation since 1995 and can house approximately 7,350 swine. Swine waste generated on the farm drain from 10 hog houses into one of two waste lagoons, which were capped in 2008 with high-density polyethylene lagoon covers and both are connected to a third covered lagoon that stores wastewater and facilitates biogas harvesting via an anaerobic digestion system. There is approximately 20.1 m (66 ft) of topographic relief from the upland area of the farm (near Sprayfield 5 in Fig. 3.1) to the lowland area of the farm (riparian buffer adjacent to “Stream” in Fig. 3.1) and elevation based on sampling location is summarized in Table 3.2. The farm is comprised of Gilead loamy sand (~54%), Blaney loamy 47 sand (~22%), and Roanoke loam (~19%) and each soil series specific to the groundwater location is listed in Table 3.2 (Appendix B) (USDA, 2022). In the lower elevation areas of the farm, the water table is shallow (within approx. 76 cm) and finer textured soils inhibit vertical infiltration. These characteristics may facilitate lateral flow within the subsurface until groundwater intersects the ground surface resulting in the formation of various groundwater seeps that emerged throughout the study area (Lin et al., 2005). Riparian buffers are present around the stream cutting through the farm; these buffers are natural areas that may attenuate plant available nitrogen (PAN) and facilitate nitrogen transformations that reduce nitrogen concentrations before reaching the stream (Lowrance et al., 1997; Christensen and Serre, 2017). Fig. 3.1 shows locations where the groundwater piezometers were installed, sprayfield locations where waste was applied, and location of the covered lagoons and anerobic digester. Table 3.2. Summary of dominant soil type, elevation, and piezometer depth based on sampling locations. Elevation Piezometer Depth Site Soil Series (m) (m) LAG NA 67.4 NA BG Blaney 63.9 2.4 SF1 Gilead 62.0 2.1 SF2 Gilead 60.1 1.5 SF3 Gilead 60.2 1.5 RB1 Roanoke 60.0 1.2 RB2 Roanoke 65.0 1.2 TP Gilead 61.1 1.7 48 Figure 3.1. Study area map showing the groundwater monitoring well locations on the farm. There is a stream that runs adjacent to the farm (SW to NE), a small seep that originates as groundwater that runs through the farm (S to N), and a small creek (east creek) that runs parallel with the seep and drains to the stream. Sprayfield sites are depicted as green polygons and labeled as “Field #.” Lagoon (LAG) is represented by a square, background piezometer (BG) is represented by a triangle, sprayfield piezometer locations (SF1, SF2, and SF3) are represented by circles, and riparian buffer piezometers (RB1 and RB2) are represented by diamonds. 3.2.2 Installation of Groundwater Piezometers Potential groundwater hotspots for nitrogen concentrations were initially identified using electrical resistivity survey data collected with an OhmMapper (Geometrics, 2020). Electrical resistivity is inversely related to electrical conductivity, which can be used as a tracer of solutes in some environments since wastewater generally contains high concentrations of ionic solutes 49 (Stewart, 2001). Piezometers were installed at the potential hotspot locations on the farm, in the riparian buffer adjacent to the stream, upgradient of the farm (background), and downgradient of sprayfields to observe nitrogen concentrations influenced by spray events at the study site (Fig. 3.2). Hand augers were used to drill to about 1.5 - 1.8 meters beneath the water table at each piezometer location. Solid PVC pipe with a 5.08 cm (2-in) diameter were coupled to well screen to construct the piezometers that were placed at depths ranging from 1.2 - 2.7 m. The piezometers were installed in the borehole and the annular space between the piezometer screen and borehole was filled with sand (gravel pack #2) until the screened portion was covered. Bentonite was used to seal the annular space above the well screen and up to the ground surface. After construction, piezometers were purged numerous times using plastic, disposable bailers. 3.2.3 Sampling Frequency and Protocols Wastewater and groundwater quality was assessed approximately monthly using a combination of in-field handheld meters and collecting samples for laboratory analysis. The study began in December 2019 and continued until January 2021. A Solinst Temperature, Level, Conductivity meter was used to measure depth to water (DTW) and temperature at each piezometer (Solinist Canada LTs, Georgetown, ON). After recording these measurements, the piezometer was purged using a disposable, PVC bailer to extract at least 2 full bailer volumes, which allowed for groundwater to recharge the piezometer. Groundwater was bailed from the piezometer and poured into the calibration cup for a YSI-556 multiparameter meter (YSI Inc., Yellow Springs, OH) to measure physical and chemical parameters including specific conductance (SC), temperature, oxidation-reduction potential (ORP), dissolved oxygen (DO), and pH. Water samples for nitrogen analysis were collected by transferring groundwater from the bailer into a clean, labeled polypropylene bottle. After sampling all groundwater piezometers, 50 a bailer was used to collect a liquid wastewater sample from an access point in the cover to measure physicochemical parameters and to collect a water sample for nutrient analysis. Samples were stored in an iced cooler and transported to East Carolina University where they were either immediately processed for analysis or frozen for processing and analysis at a future date. Water samples were processed and analyzed at the Environmental Research Laboratory at East Carolina University. Before analysis, all samples were filtered using a vacuum filtration system. Each sample was filtered through a 1.5- and 0.7-micron filter paper. After filtering, filtrate was recovered and samples were analyzed for nitrate, ammonium, TDN, and chloride using a KPM Analytics SmartChem 170 or 200 Discrete Analyzer. A Shimadzu Scientific Instruments, Inc. Total Organic Carbon (TOC) and Total Nitrogen (TN) Analyzer was used to quantify TDN concentration. A subset of sample filtrate was frozen and shipped overnight to the University of California-Davis Stable Isotope Facility for isotopic fractionation of nitrate. The isotopic fractionation of ?14N/ ?15N and ?16O/ ?18O were measured at the University of California-Davis Stable Isotope Facility using a ThermoFinnigan GasBench + PreCon trace gas concentration system interfaced to a ThermoScientific Delta V Plus isotope-ratio mass spectrometer (Bremen, Germany) (Sigman et al., 2001; Casciotti et al., 2003; Granger and Sigman, 2009). These data have been used in past studies to identify potential sources of nitrate (Showers et al., 1990; Karr et al., 2001; Silva et al., 2002; UC Davis, 2022). For example, animal or human waste typically contains elevated values of ?15N compared to other nitrate sources (e.g., fertilizers, soil organic matter, atmospheric deposition) (Hübner, 1986; Silva et al., 2002). 51 3.2.4 Data Analysis Nutrient concentrations and physicochemical parameters at various locations were compared via statistical analysis and visually represented in the R statistical framework (R Core Team, 2019). Data normality was tested using a Shapiro-Wilks test in R (Appendix H) (Shapiro and Wilk, 1965). If data exhibited normality, an analysis of variance multiple comparison test with post-hoc pairwise t-tests was used to determine if statistically significant (p < 0.05) differences existed between comparison groups. Data that was not normally distributed were tested using Kruskal-Wallis multiple comparison test with a pairwise Wilcoxon Rank Sum test if the Kruskal-Wallis test found a significant finding (Wilcoxon, 1945; Kruskal and Wallis, 1952). Groundwater flow direction was estimated using the three-point solution between SF1, SF2, and a temporary piezometer (Heath, 1983). The total head at each piezometer location and direction of slope between wells can determine the direction of groundwater flow. DTW was measured from the top of the casing and corrected for the difference between the top of the casing and land surface. Hydraulic head at each well was calculated by subtracting DTW from ground elevation. Elevation was surveyed in the field with a laser level for SF1, SF2, RB1, SF3, and TW. Elevation for BG and RB2 was estimated from a digital elevation model on ArcGIS Pro. A two-component mixing model was used to evaluate if nitrogen concentration reductions could possibly be due to dilution effects. This approach assumes that the water sampled from each well is composed of a mixture of background groundwater mixed with wastewater from the lagoon. The model was implemented using chloride concentrations from the lagoon effluent and a background well (Fig. 3.1) as end members to determine the fraction of wastewater at each sampling location using equation 1. 52 [???] ??? ? [?? ]?????? ??? = [???]?? ? [???]?? ??? = 1 ? ??? (Eq. 1) Where fgw is the fraction of the sample composed of background groundwater, [Cl -]LE is chloride concentration from lagoon effluent, [Cl-]sample is chloride concentration from the downgradient sample, [Cl-]GW is the chloride concentration from the groundwater background sample, fww is the fraction of wastewater, and fgw is the fraction of background/groundwater. The model was evaluated using chloride concentrations because chloride is assumed not to be impacted by biological or chemical removal processes like nitrogen (Li et al., 2019) and has been used extensively in other studies as a conservative tracer (e.g., Pinay et al. 1998; Humphrey et al. 2016). Beyond a uniform background value, it is also assumed that chloride is primarily contributed from wastewater sources originating from the CAFO lagoon. Mass reduction of TDN because of dilution processes only was estimated from the results of the chloride mixing model. The predicted TDN concentration resulting if only dilution occurred is given as: TDNp = fww*TDNLE + fgw*TDNGW, where TDNLE is the total dissolved nitrogen measured in the lagoon effluent and TDNGW is the total dissolved nitrogen measured in background groundwater samples that are assumed to be unaffected by the CAFO. Mass reduction efficiency (TDNe) was then calculated to assess the extent that non-conservative nitrogen removal processes occurred between the lagoon wastewater and groundwater beneath the sprayfields, riparian buffer, and background locations using equation 2. A similar equation was also used to estimate concentration reductions. ???? ? ???? ???? = ? 100% ???? 53 Where TDNe is the TDN reduction efficiency, TDNp is the predicted TDN concentration, and TDNo is the observed TDN concentration. 3.3 Results and Discussion 3.3.1. Hydrology Both precipitation and temperature tended to be greater in summer months relative to other seasons (Fig. 3.2). Precipitation was the highest during the summer (65 cm from June- August) and accounted for 33% of the total rainfall for the study period (194 cm). The annual accumulation that occurred at the site during this study (194 cm) was higher than the 30-year annual average (123 cm) (NC State Climate Office). The highest groundwater temperatures also occurred during summer months, June-August. Median temperature in warmer months (June- August) were 23?, which were 2 times greater than colder months (December-February) and this difference was statistically significant (p < 0.01). Precipitation and temperature are important for understanding groundwater hydrology because these have been identified as controlling factors leading to possibly higher or lower nitrogen concentrations (Schweigert et al., 2004; Sieling and Kage, 2006; Wick et al., 2012). Wick et al. (2012) found that higher temperatures and precipitation could decrease nitrate concentrations observed in groundwater. Increased precipitation was linked to plant growth and up take of the nutrients as well as potential for dilution. Higher temperatures were also associated with evapotranspiration associated with plant activity that potentially resulted in lower nitrate as well (Wick et al., 2012). 54 Figure 3.2. Monthly precipitation data from December 2019 to January 2021 (A) and monthly temperature of groundwater collected from piezometer locations (B) showing warmer temperatures and higher precipitation during summer months (June-August). The median DTW collected for this study varied by 84.3 cm across the site with a minimum of 20.6 cm and maximum of 104.9 cm (Table 3.3). Piezometers in the sprayfields (SF) 55 had larger DTW than the riparian buffer (RB) locations indicating that groundwater was closer to the surface in the buffer (Fig. 3.3). The BG piezometer was installed within the Blaney soil series, whereas the SF and RB piezometers were within the Gilead and Roanoke soil series, respectively. The typical soil profile for Blaney soils tends to have a greater DTW (> 203 cm) than the typical soil profile for Gilead (46-76 cm) and Roanoke (0-30 cm) soils (USDA, 2022). Furthermore, Roanoke soils tend to be poorly drained with low permeabilities, and at the current site underly surface water features and the riparian buffer region adjacent to the creek running through the farm site (USDA, 2022). DTW data can provide insight into nitrification potential due to its reliance on aerated soil, thus as DTW decreases, nitrification can be inhibited. This discussion will be explored further in a later subsection when addressing trends in groundwater nitrogen. 56 Table 3.3. Summary of median and range for physical and chemical parameters in groundwater and wastewater locations. Temp=temperature; SC=specific conductance; DO=dissolved oxygen; ORP=oxidation reduction potential; DTW=depth to water; Head=hydraulic head. LAG=Lagoon; BG=background; SF1=sprayfield 1; SF2= sprayfield 2; SF3=sprayfield 3; RB1=riparian buffer 1; RB2=riparian buffer 2. Temp. SC DO ORP DTW Head Site pH (°C) (µS/cm) (mg/L) (mV) (cm) (m) 8.3 16.4 7140 0.9 230.6 LAG NA NA (7.4-9.4) (9.1-30.2) (1137-1862) (0-3.3) (-551.0-73.4) 104.9 5.0 17.2 29 4.8 174 62.9 BG (3.3-6.4) (11.9-20.9) (21-145) (3.3-9.0) (40.0-421.9) (93.3-142.6) (61.5-62.0) 4.3 16.6 989 4.2 264.4 36.0 61.4 SF1 (3.4-6.4) (10.6-26.5) (801-1189) (2.8-5.1) (60.0-423.7) (14.3-97.2) (61.0-61.9) 4.7 16.3 232 6.3 215.8 53.2 59.6 SF2 (4.2-6.3) (10.9-25.7) (46-298) (3.8-7.1) (-8-166) (14.0-114.0) (57.8-60.0) 6.2 15.3 255 5.4 100.6 54.1 59.7 SF3 (5.9-7.2) (10.5-25.1) (208-370) (3.5-7.2) (-60.5-220.0) (21.9-94.2) (59.3-60.0) 5.4 14.6 152 3.7 95.7 20.6 59.8 RB1 (4.9-7.3) (10.6-22.9) (102-213) (3.0-4.9) (-121-166) (15.2-136.6) (58.6-59.8) 4.8 15.1 205 3.7 201.6 20.6 64.8 RB2 (4.1-6.5) (11.4-22.6) (176-252) (0.3-10.9) (35.0-260.0) (17.7-42.4) (64.4-64.6) 59.6 TP NA NA NA NA NA NA (59.6-60.3) 57 Figure 3.3. Depth to water (DTW) for piezometer locations collected monthly during the study. The depth to the water table at the site varied substantially throughout the year (Appendix D) (Fig. 3.4). Greater values of DTW, which indicates a deeper water table, occurred during the warmer months. The water table varied by around 100 cm at SF locations, whereas it varied by around 50cm at BG piezometer and about 120 in RB locations. Fig. 3.2 shows that the warmer months had the highest precipitation (Fig. 3.2 A) and higher temperatures (Fig. 3.2 B). The deeper water tables occurring during summer (June-August) despite the higher precipitation is likely due to high evapotranspiration rates in this season. The coastal plain loses about 70% of annual precipitation to evapotranspiration according to Sun et al. (2002) and O’Driscoll (2012). A study by Chvosta et al (2004) used various monitoring locations in eastern North Carolina near swine farm operation and found that 50 – 100% precipitation was lost to evapotranspiration. 58 Evapotranspiration in the warmer months typically outweigh precipitation and result in lower hydraulic heads (Harden and Spruill, 2008; O’Driscoll, 2012). Figure 3.4. Time series of depth to water (DTW) at piezometer locations throughout the study. Locations are Background (BG), Sprayfield 1 (SF1), Sprayfield 2 (SF2), Sprayfield 3 (SF3), Riparian Buffer 1 (RB1), and Riparian Buffer 2 (RB2). Groundwater piezometer locations ranged from approximately 59 - 65 m of elevation above sea level (Table 3.2). Initially, SF1, SF2, and RB1 were installed in a transect to assess groundwater quality with distance from the sprayfield. Since the depth to water at these piezometers was also monitored, the hydraulic heads can provide insight on groundwater flow direction as groundwater flows in the direction from higher hydraulic head towards lower hydraulic head (Post and von Asmuth, 2013). However, the orientation of these piezometers did not allow for determination of groundwater flow direction. Therefore, a new temporary 59 piezometer (TP) was also installed in February 2021 to determine groundwater flow direction using the three-point solution with SF1 and SF2 (Heath, 1983). The hydraulic head medians ranged from 59.4 - 64.8 m and an overall range of 57.5 - 64.6 m (Table 3.3). SF1 had the highest hydraulic head (60.9 m) relative to the other piezometers installed in the sprayfield. The median hydraulic head value for SF2, RB1, and TP was 59.8, 59.4, and 59.6 m, respectively. Thus, there was a small gradient of hydraulic head between these piezometers, but the direction of groundwater flow likely fluctuated between RB1 (northwestwardly) and TP (northeastwardly) throughout the duration of the study. This observation was supported via direction observation of 2 seeps, which included a groundwater seep that fed into a channel that flowed perennially (seep labeled on Fig. 3.1) and an ephemeral seep that only surfaced during wet months (from late October to late March). The latter of which surfaced underneath the blue arrow on Fig. 3.3. Based on hydraulic head values in March 2021 and May 2021 from SF1, RB1, and TP, groundwater flow direction is likely flowing north/northeast (Fig. 3.5). 60 Figure 3.5. Sprayfield 4 piezometer locations, average groundwater flow direction, and average hydraulic heads based on March 2021 and May 2021 indicating north/northeast GW flow direction. 3.3. Specific Conductivity and Geophysical Data Specific conductivity (SC) data observed for water samples at each piezometer suggest that groundwater beneath and downgradient of sprayfields are influenced by land application of wastewater. Wastewater in the lagoon contained the highest median value of SC (7140 µS/cm), which was up to approximately 31, 47, and 246 times greater than median SC in SF, RB, and BG piezometers (Fig. 3.6). Furthermore, these differences were statistically significant (p < 0.01). SF piezometers contained median SC values that ranged from 232 (SF2) to 989 (SF1) µS/cm. These values were elevated relative to median SC in BG groundwater by a factor of about 8, 9, and 34 times for SF3, SF2, and SF1, respectively, and differences were statistically significant (p < 0.01). RB piezometers contained median SC values that ranged from 152 (RB1) to 205 (RB2) µS/cm. These values were elevated relative to median SC in BG groundwater by a factor of about 5 and 7 times for RB1 and RB2, respectively, and differences were statistically different (p < 0.01). 61 Figure 3.6. Boxplot of specific conductance of groundwater and wastewater from the lagoon (LAG). BG= background, SF= sprayfield, and RB= riparian buffer. SC data indicated that groundwater piezometers within and downgradient of sprayfields were affected by wastewater applications. SF1 consistently contained the highest SC values in the groundwater when compared to other groundwater monitoring locations. Piezometers SF2, SF3, RB1, and RB2 are located on the periphery (SF) or downgradient (RB) of sprayfields, thus SC values were lower than SF1, which was located in the center of a sprayfield (Fig. 3.1). Fig. 3.7 shows variability of resistivity within the sprayfield, which is the inverse of specific conductivity (Heaney, 2003). A study by Michalopoulos et al. (2014) found specific conductivity values in groundwater sampled from monitoring wells adjacent to a swine CAFO ranged from 814 - 1,965 µS/cm. SF1 was the only piezometer to fall within this range further suggesting waste application location is primarily impacting SF1 (Table 3.3). However, Michalopoulos et al. 62 (2014) studied a larger CAFO that housed about 16,000 swine, which was more than double the number of swine at the current study (7,350). Figure 3.7. OhmMapper data showing potential hotspots (circled) of low resistivity/high conductivity for well installation. This figure was created by Michael O’Driscoll. 3.3.2 Nitrogen Concentrations Wastewater in the lagoon contained the highest median concentration of TDN (485 mg/L), which was about 97, 72, and 1,155 times greater than median values of SF, RB, and BG, respectively (Table 3.4; Fig. 3.8). These differences in TDN were significant (p < 0.01). TDN concentrations in the lagoon ranged from 304.8 to 1824.9 mg/L. Past studies found that TKN ranges of 276 - 1720 mg/L and 564 - 1783 mg/L for open and capped lagoons, respectively 63 (Miner et al., 2003; VanderZaag et al., 2010; Ducey et al., 2019). These values may be variable due to differences in the number of swine housed by the farms. Ducey et al. (2019) studied two farms, one with an uncovered lagoon that houses 2,100 - 2,200 swine and a second with a covered lagoon that that housed 1,200 - 1,500. Miner et al. (2003) studied a swine farm containing 1,224 swine and VanderZaag et al. (2010) studied cow manure from an unknown number of animals. Ducey (2019) found that open and closed lagoons had a mean TKN of 473 mg/L (range: 276 - 630) and 1009 mg/L (range: 906 - 1110), respectively. Overall, studies have shown that lagoon TDN and TKN concentrations are extremely variable. In the current study, ammonium was the most dominant nitrogen species in lagoons contributing 74% of TDN (Table 3.4). This was expected since animal waste consists predominantly of organic nitrogen and ammonium species; furthermore, mineralization of organic nitrogen to ammonium can occur in anaerobic environments (e.g., waste lagoon) (Burns et al., 1987; DeSutter et al., 2005; Harden, 2015). Groundwater quality data suggest that land application of wastewater from a capped lagoon can increase nitrogen concentrations in groundwater beneath and adjacent to sprayfields. SF piezometers contained median TDN concentrations that ranged from 1.7 mg/L (SF2) to 27.2 mg/L (SF1). The TDN observed in groundwater under the sprayfield was substantially lower than that in the waste lagoon (485 mg/L), with reductions of 94.4% (SF1) to 99.6% (SF2). However, median TDN concentrations in the sprayfield were elevated compared to background (BG) by a factor of 181, 11, and 22 times for SF1, SF2, and SF3, respectively (Table 3.4). These differences were statistically significant at p < 0.01, except for SF2 which was significant at p= 0.05. RB piezometers contained median TDN concentrations of 2.9 mg/L (RB1) and 9.1 mg/L (RB2). These values were elevated compared to median TDN concentrations in piezometer BG 64 by about 19 and 61 times for RB1 and RB2, respectively, and differences were significant (p < 0.01). Despite high concentration reductions, TDN concentrations in RB piezometers remained elevated compared to BG groundwater. Stone et al. (1998a), Karr et al. (2001), and Spruill et al. (2002) found nitrate concentrations in groundwater at swine farms ranged from 0.15 - 35 mg/L. TDN median concentrations in all piezometers fell within this range. Overall, studies show that TDN concentrations are elevated in groundwater at swine CAFOs (Stone et al., 1998a; Karr et al., 2001; Spruill et al., 2002). 65 Table 3.4. Summary of median and range (parenthesis) concentrations for ammonium (NH +4 ), nitrate (NO -3 ), total dissolved nitrogen (TDN), and chloride (Cl-). NH + -4 NO3 TDN Cl- Site (mg/L) (mg/L) (mg/L) (mg/L) 484.7 LAG 359.0 0.15 674.3 (304.8-1824.9) (269.8-1824.5) (0.06-0.95) (499.6-903.4) BG 0.03 0.06 0.42 3.88 (0.02-0.49) (0.03-0.18) (0.26-1.53) (2.71-5.28) SF1 9.59 17.9 27.2 229.8 (6.33-11.2) (8.49-34.0) (16.6-43.5) (193.8-245.0) SF2 0.03 1.19 1.67 14.7 (0.01-0.73) (0.80-3.32) (0.86-3.36) (6.94-23.3) SF3 0.05 4.9 5.02 6.18 (0.01-0.17) (0.03-15.0) (1.10-14.6) (2.55-22.8) RB1 0.89 1.29 2.85 18.0 (0.48-1.69) (0.02-10.1) (2.03-10.8) (12.2-36.0) RB2 0.04 8.98 9.10 33.8 (0.01-0.21) (4.00-12.0) (5.10-12.1) (28.7-44.2) 66 Figure 3.8. Concentrations of total dissolved nitrogen, nitrate, and ammonium in the lagoon and groundwater piezometers. Nitrogen speciation data suggested that sprayfields can increase availability of plant available nitrogen (PAN) species in groundwater. The majority of the TDN in groundwater consisted of nitrate and ammonium (Table 3.4; Fig. 3.8). Nitrate accounted for >65% of median TDN concentrations in groundwater collected from the SF piezometers and RB2. Groundwater at RB1 was also mostly nitrate (45% of median TDN), but elevated percentages of ammonium (31% of median TDN) and dissolved organic nitrogen (DON) (24% of median TDN) were also observed. Studies like Stone et al. (1998a), Karr et al. (2001), and Israel et al. (2005) also found nitrogen speciation to be mostly nitrate in groundwater beneath sprayfields at CAFOs. Ammonium tended to comprise < 2% of median TDN concentrations in groundwater from SF and RB piezometers, apart from SF1 (35% of TDN) and RB1 (31% of TDN). The elevated ammonium percentages at SF1 and RB1 suggest that nitrification may be inhibited from 67 biogeochemical processes or transport pathways between the source and sampling location may be shorter. Ammonium can leach to groundwater if the cation exchange sites are at capacity or the DTW is shallow and does not allow for complete nitrification (Bouwer et al., 1980). Collectively, PAN species accounted for >73% of TDN at groundwater monitoring locations, except for the BG piezometer. BG groundwater consisted of mostly DON (79%), followed by 14% nitrate and 7% ammonium of TDN. Results from this study and others (Stone et al., 1998a; Karr et al., 2001; Spruill et al., 2002; Israel et al., 2005) suggest that land application of swine wastewater can be a substantial source of nitrogen. Riparian buffers are used near sprayfields and surface water to remove excess nitrate through denitrification, plant uptake, and immobilization (Mayer et al., 2007). The elevated nitrate in RB1 and RB2 indicate complete denitrification may not be occurring. Inadequate treatment may be due to shallow DTW may not be allowing enough time for denitrification, or the buffer may not be able to handle the amount of excess nitrate (Bouwer et al., 1980). Israel et al. (2005) riparian buffer widths were 41 to 87 m and reported average nitrate concentrations in riparian buffer locations ranged from <0.5 to 15 mg/L. RB1 and RB2 TDN concentrations ranged from 2.03 - 12.1 mg/L falling within this range 100% of the time. RB1 showed an 89% reduction from nearby sprayfield piezometer, SF1, but elevated concentrations compared to BG may indicate compete denitrification is not occurring. 3.3.3. Mass Reduction Estimates Concentration reductions of TDN between wastewater and groundwater locations could be due to denitrification, plant and/or microbial uptake, anaerobic ammonium oxidation, volatilization, or dilution. All of which are processes that reduce the mass of nitrogen except for dilution. Chloride/TDN ratios and a two-component mixing model provides insight regarding the 68 extent to which dilution occurred. If dilution was the dominant mechanism of concentration reductions, then the ratio of chloride to TDN would be similar between wastewater and groundwater (Lowrance, 1992). Wastewater in the lagoon had a chloride/TDN ratio of 1.53, which was lower than groundwater from SF and RB piezometers except for SF3 (Table 3.5). Chloride/TDN values in groundwater at SF1, SF2, RB1, and RB2 were 8.44, 8.83, 6.30, and 3.72, respectively. Groundwater ratios were approximately 2 – 6 times greater than the chloride/TDN ratio in wastewater, indicating that dilution was not likely the only mechanism reducing nitrogen concentrations. The predicted TDN in Table 3.5 is an estimate that assumes dilution is the only reduction mechanism calculated from the binary mixing model by Equation 1. The observed TDN was substantially lower than these predicted values for all groundwater piezometers except for SF3. Therefore, the difference between the two was estimated to account for mass reductions of TDN due to other biogeochemical processes. Mass reductions ranged from -161% – 82% in groundwater at SF and RB piezometers. The mass reduction estimate for SF3 was not likely accurate, which is discussed further in the next paragraph. Thus, if removing this observation, mass reductions of TDN ranged from 55 – 82%. 69 Table 3.5. Two-component mixing model using chloride and TDN to estimate TDN mass reductions. Chloride concentrations (Cl-), fraction of wastewater (fraction of WW), fraction of groundwater (fraction of BG), predicted total dissolved nitrogen (predicted TDN), observed total dissolved nitrogen (observed TDN), chloride/total dissolved nitrogen ratios (Cl-/TDN), and total dissolved nitrogen mass reduction (TDN mass reduction) are shown for sites Lagoon (LAG), Sprayfield 1 (SF1), Sprayfield 2 (SF2), Sprayfield 3 (SF3), Riparian Buffer 1 (RB1), Riparian Buffer 2 (RB2), and Background (BG). SF3 shows a negative mass reduction, indicating that this location is a source of TDN. However, this is unlikely and due to limitations of the mixing model estimate since SF3 Cl- and BG Cl- is so similar. Site Location Cl- Fraction Fraction Predicted Observed Cl-/TDN TDN Mass (mg/L) of WW of GW TDN TDN Reduction (mg/L) (mg/L) LAG Lagoon 743.33 1.00 0.00 484.72 1.53 SF1 Sprayfield 229.84 0.31 0.69 148.41 27.25 8.44 82% SF2 Sprayfield 14.71 0.01 0.99 7.51 1.67 8.83 78% SF3 Sprayfield 6.18 0.00 1.00 1.92 5.02 1.23 -161% RB1 Buffer 17.96 0.02 0.98 9.64 2.85 6.30 70% RB2 Buffer 33.84 0.04 0.96 20.04 9.10 3.72 55% BG Background 3.88 0.00 1.00 0.42 9.24 The highest TDN mass reduction was observed in groundwater at SF1. Despite high mass reduction estimates, TDN concentrations found in groundwater at SF1 were substantially elevated with a median TDN value of 27.3 mg/L (Table 3.5). TDN concentrations in groundwater downgradient and adjacent to sprayfields (e.g., SF2, SF3, and RB piezometers) suggest that further treatment can occur in the subsurface before reaching surface waters. However, these concentrations remain elevated relative to the US EPA’s nutrient reference criteria for Ecoregion IX (0.7 - 1.0 mg/L of total nitrogen) (US EPA, 2000). If these concentrations were to reach surface waters (especially nutrient-sensitive waters), it could contribute to negative environmental impacts like eutrophication, algal blooms, and fish kills (Vitousek et al. 1997; Tilman et al. 2002; Townsend et al. 2003) or human health impacts like methemoglobinemia (or “blue baby syndrome”) and cancer (Ward et al., 2005). One of the limitations with the two-component mixing model is the potential to over- or underestimate mass reductions when chloride concentrations vary across several orders of 70 magnitude. For example, the model indicates that TDN concentrations at SF3 should be substantially lower than what was observed (Table 3.5). Thus, mass reduction estimates suggest that SF3 is a source of TDN, which is not likely. Since chloride concentrations at SF3 were more like BG groundwater than the wastewater in the lagoon, the fraction of wastewater was minimal (0.019%) and predicted that 99.6% of the TDN was diluted. If this was the case, chloride concentrations should be approximately 3 mg/L. However, groundwater at SF3 contained median chloride and TDN concentrations that were approximately 1.5 and 12 times greater than BG, respectively (Table 3.5). This mixing model was based on the median TDN and chloride concentrations, so one way to better understand these estimates is to explore the range in concentrations for TDN and chloride (Appendix I). Based on minimum TDN and chloride concentrations, TDN concentrations reductions range from -579%-86%, with SF3 still resulting in a negative TDN mass reduction for reasons previous discussed. However, the maximum TDN and chloride concentrations from the study result in a range of 61%-91%. SF3 has a TDN concentration reduction of 61% based on maximum TDN and chloride concentrations. The maximum concentrations helped differentiate between the background concentrations and SF3 to help with the underestimating that was likely happening based on the median, however, the maximum concentrations may now be overestimating the reductions. 3.3.4 Sources and Pathways of TDN Isotopic fractionation of nitrate data suggested that animal and/or human waste was the most likely source of nitrate in groundwater at all sampling locations (Fig. 3.9). The enriched ?15N and ?18O values of nitrate in groundwater suggest denitrification is occurring at all sampling locations. Denitrification occurs in anaerobic environments through denitrifying bacteria that will use nitrate with the lighter nitrogen isotope (14N) first, thereby increasing values of the 71 heavier nitrate isotope (15N) (Silva et al., 2002). Thus, these data are consistent with waters originating as human or animal waste with isotopic signatures that have been modified through denitrification, thereby suggesting the CAFO as the most probable source. Figure 3.9. Stable isotope data indicating possible nitrate sources showing boundaries of possible nitrate sources derived by previous studies (Kendall and McDonnell, 1998; Silva et al., 2002). NP=nitrate in precipitation, SNF= synthetic nitrate fertilizer, AF= ammonium fertilizer, SN=synthetic nitrogen, AHW=animal or human waste. NP values extend to 60 ‰ ?18O but this range was truncated for this figure. Land cover data also support this interpretation that the most likely source of nitrate in groundwater originated from swine wastewater. SF piezometers were installed within sprayfields receiving waste application and RB piezometers were installed in riparian buffers adjacent to sprayfields where waste is applied. No sources of human waste, such as septic tanks, are known to be in areas that could contribute to groundwater at these sites. Wildlife (e.g., deer, rabbits, 72 snakes, alligators, birds, etc.) was commonly observed at the field site by researchers, thus their defecate could be a potential source of nitrate if ammonification and nitrification of organic nitrogen occurs (Panno et al., 2006), but these contributions are likely to be small and captured within the background data. 3.4 Conclusion The goal of this study was to evaluate groundwater quality impacts downgradient of a swine CAFO using a capped lagoon system. The median concentration of TDN found in the lagoon (485 mg/L) fell within ranges found in other studies, but overall indicate that TDN of waste stored in lagoons is highly variable (276 - 1720 mg/L for open lagoons). Shallow groundwater beneath and adjacent to sprayfields contained a median TDN concentration that was up to 99% lower than median TDN in wastewater. Despite these reductions, sprayfield (median: 5.0 mg/L) and riparian buffer (median: 6.2 mg/L) piezometers had elevated TDN concentrations compared to the median TDN concentrations of the background well (0.42 mg/L) outside of the region of influence of the farm and typical background nitrate levels established by Mueller and Helsel, 1996 (2 mg/L) (Table 3.1). A two-component mixing model suggested that TDN mass reductions ranged from 55 – 82% more than can be explained by dilution alone. Thus, biogeochemical mass reduction mechanisms likely accounted for substantial reductions of TDN observed at the site by removing ammonium and nitrate. Furthermore, isotopic analysis showed that groundwater contained elevated values of ?15N and ?18O, which also supports that denitrification accounted for reductions in TDN by way of removing nitrate. 73 CHAPTER 4: CONCLUSION AND MANAGEMENT IMPLICATIONS 4.1. Groundwater-Surface Water Interactions Chapters 2 and 3 investigated the surface water and groundwater quality impacts at the farm separately, however, understanding how surface water and groundwater interact is also important for future management implications for CAFOs with capped lagoon systems. Studies like Böhlke (2002) and Howarth (2002) have found that contamination in groundwater beneath sprayfields is the main contributor to eutrophication in some watersheds. Understanding the sources and pathways of nutrients, in this case nitrogen species, is important for nutrient management. Water quantity and quality may be directly related to surface water and groundwater interactions at the CAFO. Normalized TDN exports showed an increase of 1121% in the creek when comparing sampling locations upstream and downstream of the farm (i.e., UP and DOW). The seep was identified as a potential conduit of excess nutrients released by the farm to the nearby stream. Based on nitrogen exports, the seep contributed a median 25% of TDN observed in the creek downstream of the farm (DOW) with a range of 6%-33% during most sampling events (8 out of 10 sampling events). Although groundwater TDN exports were not calculated for this study, the elevated TDN concentrations found in the groundwater and variable TDN exports of the seep indicate that groundwater is likely the other pathway for TDN to enter the surface water system. Elevated TDN concentrations were found beneath sprayfields and riparian buffers. If the seep is exporting a median TDN export of 25%, the groundwater may be contributing up to 75% of the excess TDN found downstream of the farm. Assuming a simple nutrient mass budget, it was estimated that groundwater contributed approximately 5, 700 kg/yr. This estimate was derived by taking the difference in the median TDN load between DOW, SD, and UP (DOW [7,818.4 kg/yr] – SD [1,804.7 kg/yr] – UP [316.7 kg/yr]). Riparian buffers are present along the stream adjacent to the farm and are typical best management practices suggested to help offset inputs of nitrogen that can result in water quality degradation. However, groundwater TDN concentrations within the riparian buffers (RB1 and RB2) for this study showed elevated concentrations ranging from 2.03 mg/L to 12.1 mg/L. There are TDN concentration reductions between SF1 (27.2 mg/L) and the nearby riparian buffer location, RB1 (2.85 mg/L); however, there are still elevated TDN concentrations in RB1 compared to BG (0.42 mg/L). Elevated TDN concentrations and exports were also still found downstream of the farm, despite riparian buffers present. Elevated TDN concentrations in the RB and DOW compared to BG and UP indicate the riparian buffer is not completely denitrifying or retaining the excess nitrate from waste application to the sprayfields before discharging into the nearby stream. A study by Sloan et al. (1999) saw similar findings where downstream of a CAFO was elevated compared to the UP with elevated nitrate levels in groundwater beneath the nearby spray field, thus concluding that the nitrate from spray events was entering the stream through the groundwater. Adjacent groundwater was as high as 40 mg/L with downstream surface water samples up to 7 mg/L, indicating denitrification was likely taking place but the riparian buffer was not fully removing the nitrate and therefore discharging excess nitrate to the stream. Another study by Karr et al. (2001) suggested that excess nitrate being discharge via groundwater could be due to shallow groundwater tables that did not have enough time to denitrify in the riparian buffer or that the riparian buffer was not capable of denitrifying that much excess nitrate. Puckett (2004) also suggests that some of the nitrate rich groundwater can flow under the buffer without full treatment in areas with ditches. 75 5.2 Management Implications This study shows swine CAFO waste management can be a source of excess nitrogen species in groundwater and nearby surface water. Nitrogen species have been cited to cause eutrophication and fish kills in nutrient sensitive watersheds so understanding and quantifying nitrogen exports from places like swine CAFOs is important for nutrient management. Based on elevated nitrogen species in groundwater below sprayfields, within the seep, and being exported downstream, there are some strategies that may be implemented to reduce nitrogen species in potentially nutrient sensitive watersheds. The Cape Fear Watershed contains almost half of the swine population in North Carolina (Mallin et al., 2015) and has streams that are nutrient sensitive, and streams used for water supply, recreation, habitat, and drain into estuaries and other marine habitats used for shellfish harvesting. This farm does not immediately connect to a nutrient sensitive stream, but there are other streams in the watershed that are nearby or downstream of CAFOs that these management implications are applicable (NC DEQ). Studies like Bradford et al. (2008) and Christenson and Serre (2017) indicated that current nutrient management plans may be too general and not be completely effective at estimating the amount of nitrogen to be completely utilized by the crops. This study along with other studies showed that lagoon TDN was variable and further support that nutrient management plans may be too general. Another study by Rosov et al. (2020) also challenged the effectiveness of current management plans and suggested that sampling of the waste and soils should occur more frequently. Nutrient management plans could be more specific to the farm and evaluate the needs of the farm more frequently to mitigate excess nitrogen species in water resources (Bradford et al., 2008; Christiansen and Serre, 2017). 76 Other technologies and best management practices may also be effective combined with more specific and upgraded nutrient management plans. Lagoon waste could be treated using advanced wastewater techniques (e.g., biological nutrient removal) that remove a portion of labile nutrients prior to irrigation or land application. Reducing some of the labile nutrients from swine waste may mitigate offsite transport of nutrients, especially more mobile species such as nitrate. Best management practices like subsurface bioreactors, instream bioreactors, or wetlands could be implemented to better facilitate nitrogen removal in surface water and groundwater. Instream and subsurface bioreactors have been shown to effectively promote denitrification in agricultural studies using additional carbon source (wood chips). Another way to promote denitrification is to increase residence time with controlled drainage paired with the bioreactors (Woli et al., 2010). Management practices like bioreactors are also valuable options because this technology is cost efficient and can be scaled for various flow sizes without using additional land area (Robertson and Merkley, 2009). Wetlands may improve water quality by removing nitrogen through denitrification and immobilization, however this would require additional land/sprayfield area (Poe et al., 2003). Based on this, subsurface bioreactors, instream bioreactors, and wetlands are all potential ways to improve water quality and reduce nitrogen inputs. There is a potential area in Field 3 for a constructed wetland, however limitations to this would be losing sprayfield area. The most feasible strategy for cost and land area would be bioreactors with controlled drainages in the stream and seep. 77 REFERENCES Amini, A., Aponte-Morales, V., Wang, M., Dilbeck, M., Lahav, O., Zhang, Q., Cunningham, J. A., and Ergas, S. J. (2017). Cost-effective treatment of swine wastes through recovery of energy and nutrients. Waste Management, 69, 508–517. Behnke, J. (1975). A summary of the biogeochemistry of nitrogen compounds in ground water. Journal of Hydrology, 27(1-2), 155–167. Birch, A. L., R. E. Emanuel, A. L. James, and E. G. Nichols (2016). Hydrologic Impacts of Municipal Wastewater Irrigation to a Temperate Forest Watershed." Journal of Environmental Quality 45(4): 1303-1312. Böhlke, J. K. (2002). Groundwater recharge and agricultural contamination. Hydrogeology Journal, 10(1), 153-179. Bouwer, H., Rice, R. C., Lance, J. C., and Gilbert, R. G. (1980). Rapid-infiltration research at flushing meadows project, Arizona. Water Pollution Control Federation, 2457-2470. Bradford, S. A., Segal, E., Zheng, W., Wang, Q., and Hutchins, S. R. (2008). Reuse of concentrated animal feeding operation wastewater on agricultural lands. Journal of Environmental Quality, 37(S5). Brown, C. N., Mallin, M. A., and Loh, A. N. (2020). Tracing nutrient pollution from industrialized animal production in a large coastal watershed. Environmental Monitoring and Assessment, 192(8).?? Burkholder, J. A., Libra, B., Weyer, P., Heathcote, S., Kolpin, D., Thorne, P. S., and Wichman, M. (2007). Impacts of waste from concentrated animal feeding operations on water quality. Environmental Health Perspectives, 115(2), 308–312. Burns, J. C., Westerman, P. W., King, L. D., Overcash, M. R. and Cummings, G. A. (1987) Swine manure and lagoon effluent applied to a temperate forage mixture: I. Persistence, yield, quality, and elemental removal. Journal of Environmental Quality 16, 99–105. Cahoon, L. B., Mikucki, J. A., and Mallin, M. A. (1998). Nitrogen and phosphorus imports to the Cape Fear and Neuse River basins to support intensive livestock production. Environmental Science and Technology, 33(3), 410–415. Casciotti, K. L., Sigman, D. M., Hastings, M. G., Böhlke, J. K., and Hilkert, A. (2002). Measurement of the oxygen isotopic composition of nitrate in seawater and freshwater using the denitrifier method. Analytical Chemistry, 74(19), 4905–4912. Castro, M. S., Driscoll, C. T., Jordan, T. E., Reay, W. G., and Boynton, W. R. (2003). Sources of nitrogen to estuaries in the United States. Estuaries, 26(3), 803–814. Cerling, T. E., Hart, J. A., and Hart, T. B. (2004). Stable isotope ecology in the Ituri forest. Oecologia, 138(1), 5–12. Cheng, F. Y., and Basu, N. B. (2017). Biogeochemical hotspots: Role of small water bodies in landscape nutrient processing. Water Resources Research, 53(6), 5038–5056. Christenson, E. C., and Serre, M. L. (2017). Integrating remote sensing with nutrient management plans to calculate nitrogen parameters for swine CAFOs at the sprayfield and sub-watershed scales. Science of The Total Environment, 580, 865–872. Chvosta, J., K. Zering, and B. Norwood. (2004). Blame it on the Weather: Cost and Design of Manure Management under Extreme Weather Conditions on North Carolina Swine Farms. 2004 Annual Meeting, February 18, Tulsa, OK. Southern Agricultural Economics Association. Deal, S. C., Gilliam, J. W., Skaggs, R. W., and Konyha, K. D. (1986). Prediction of nitrogen and phosphorus losses as related to agricultural drainage system design. Agriculture, ecosystems & environment, 18(1), 37-51. Del Rosario, K. L., Humphrey, C. P., Mitra, S., and O'Driscoll, M. A. (2013). Nitrogen and carbon dynamics beneath on-site wastewater treatment systems in Pitt County, North Carolina. Water Science and Technology, 69(3), 663–671. DeSimone, L. A., and Howes, B. L. (1996). Denitrification and nitrogen transport in a coastal aquifer receiving wastewater discharge. Environmental Science and Technology, 30(4), 1152–1162. DeSutter, T. M., Pierzynski, G. M., and Ham, J. M. (2005). Movement of lagoon-liquor constituents below four animal-waste lagoons. Journal of Environmental Quality, 34(4), 1234–1242. Ducey, T. F., Rashash, D. M., and Szogi, A. A. (2019). Differences in microbial communities and pathogen survival between a covered and uncovered anaerobic lagoon. Environments, 6(8), 91. Dobriyal, P., Badola, R., Tuboi, C., and Hussain, S. A. (2016). A review of methods for monitoring streamflow for Sustainable Water Resource Management. Applied Water Science, 7(6), 2617–2628. Dodds, W. K., Bouska, W. W., Eitzmann, J. L., Pilger, T. J., Pitts, K. L., Riley, A. J., Schloesser, J. T., and Thornbrugh, D. J. (2008). Eutrophication of U.S. freshwaters: Analysis of potential economic damages. Environmental Science and Technology, 43(1), 12–19. Donner, S. D., Kucharik, C. J., and Foley, J. A. (2004). Impact of changing land use practices on nitrate export by the Mississippi River. Global Biogeochemical Cycles, 18(1). 79 Evans, R.O., Westerman, P.W., and Overcash, M.R. (1984). Subsurface drainage water quality from land application of swine lagoon effluent. Transactions of the American Society of Agricultural Engineers, 27(2), 473–478. Granger, J., and Sigman, D. M. (2009). Removal of nitrite with sulfamic acid for nitrate N and O isotope analysis with the denitrifier method. Rapid Communications in Mass Spectrometry, 23(23), 3753–3762. Harden, S. L., and Spruill, T. B. (2008). Factors affecting nitrate delivery to streams from shallow ground water in the North Carolina coastal plain. Scientific Investigations Report. Sun et al 2002 Harden, S. L. (2015). Surface-water quality in agricultural watersheds of the North Carolina Coastal Plain associated with concentrated animal feeding operations.?Scientific Investigations Report, 1-55, 1-13.?? Heaney, M. (2003). Electrical conductivity and resistivity. Electrical Measurement, Signal Processing, and Displays. Heath, R. C. (1983). Basic ground-water hydrology. U.S. G.P.O. Howarth, R. W., Sharpley, A., and Walker, D. (2002). Sources of nutrient pollution to coastal waters in the United States: Implications for achieving coastal water quality goals. Estuaries, 25(4), 656-676. Hribar, C. (2010). Understanding Concentrated Animal Feeding Operations and Their Impact on Communities. Accessed September 30, 2022. Retrieved from https://www.cdc.gov/nceh/ehs/docs/understanding_cafos_nalboh.pdf Huang, J., Borchardt, D., & Rode, M. (2022). How do inorganic nitrogen processing pathways change quantitatively at daily, seasonal, and multiannual scales in a large agricultural stream? Hydrology and Earth System Sciences, 26(22), 5817-5833. Hübner, H. (1986). Isotope effects of nitrogen in the soil and biosphere. In: Fritz, P. and Fontes, J.C., (eds.), Handbook of Environmental Isotope Geochemistry, Vol. 2B. Elsevier, Amsterdam, p. 361–425. Humphrey, C. P., O'Driscoll, M. A., and Zarate, M. A. (2010). Controls on groundwater nitrogen contributions from on-site wastewater systems in coastal North Carolina. Water Science and Technology, 62(6), 1448–1455. Humphrey, C. P., Jernigan, J., Iverson, G., Serozi, B., O’Driscoll, M., Pradhan, S., and Bean, E. (2016). Field evaluation of nitrogen treatment by conventional and single-pass sand filter onsite wastewater systems in the North Carolina Piedmont. Water, Air, and Soil Pollution, 227(8). 80 Husk, B. R., Anderson, B. C., Whalen, J. K., and Sanchez, J. S. (2017). Reducing nitrogen contamination from agricultural subsurface drainage with denitrification bioreactors and controlled drainage. Biosystems Engineering, 153, 52-62. Israel, D. W., Showers, W. J., Fountain, M., and Fountain, J. (2005). Nitrate movement in shallow ground water from swine-lagoon-effluent spray fields managed under current application regulations. Journal of Environmental Quality, 34(5), 1828–1842. Jordan, T. E., Correll, D. L., & Weller, D. E. (1997). Effects of agriculture on discharges of nutrients from coastal plain watersheds of Chesapeake Bay. Journal of Environmental Quality, 26(3), 836–848. Karr, J. D., Showers, W. J., Gilliam, J. W., and Andres, A. S. (2001). Tracing Nitrate Transport and environmental impact from intensive swine farming using Delta nitrogen-15. Journal of Environmental Quality, 30(4), 1163–1175. Katz, B. G., Griffin, D. W., and Davis, J. H. (2009). Groundwater quality impacts from the land application of treated municipal wastewater in a large karstic spring basin: Chemical and microbiological indicators. Science of the Total Environment, 407(8), 2872-2886. Kendall, C., and McDonnell, J. J. (1998). Isotope tracers in catchment hydrology (pp. 531–552). Amsterdam: Elsevier Science. Kester, R. A., De Boer, W., and Laanbroek, H. J. (1997). Production of no and N(INF2)O by pure cultures of nitrifying and denitrifying bacteria during changes in aeration. Applied and Environmental Microbiology, 63(10), 3872–3877. Kruskal, W.H. and Wallis, W.A. (1952) Use of Ranks in One-Criterion Variance Analysis. Journal of the American Statistical Association, 47:260, 583-621. Li, C., Li, B., and Bi, E. (2019). Characteristics of hydrochemistry and nitrogen behavior under long-term managed aquifer recharge with reclaimed water: A case study in North China. Science of The Total Environment, 668, 1030–1037. Lin, H., Bouma, J., Wilding, L. P., Richardson, J. L., Kutílek, M., and Nielsen, D. R. (2005). Advances in hydropedology. Advances in Agronomy, 1–89. Lowrance, R., Todd, R., Fail Jr, J., Hendrickson Jr, O., Leonard, R., & Asmussen, L. (1984). Riparian forests as nutrient filters in agricultural watersheds. BioScience, 34(6), 374-377. Lowrance, R. (1992). Groundwater nitrate and denitrification in a coastal plain riparian forest. Journal of Environmental Quality, 21(3), 401–405. Lowrance, R., Altier, L. S., Newbold, J. D., Schnabel, R. R., Groffman, P. M., Denver, J. M., Correll, D. L., Gilliam, J. W., Robinson, J. L., Brinsfield, R. B., Staver, K. W., Lucas, W., and Todd, A. H. (1997). Water quality functions of riparian forest buffers in Chesapeake Bay Watersheds. Environmental Management, 21(5), 687–712. 81 Lu, M., Lamichhane, P., Liang, F., Imerman, E., and Chai, M. (2007). Identification of odor causing compounds in a commercial dairy farm. Water, Air, and Soil Pollution: Focus, 8(3-4), 359–367. Mallin, M. A., and Cahoon, L. B. (2003). Industrialized animal production - A major source of nutrient and microbial pollution to aquatic ecosystems. Population and Environment, 24(5), 369–386.? Mallin, M. A., McIver, M. R., Ensign, S. H., and Cahoon, L. B. (2004). Photosynthetic and heterotrophic impacts of nutrient loading to blackwater streams. Ecological Applications, 14(3), 823–838. Mallin, M. A., McIver, M. R.,?Robuck, A. R., and Dickens, A. K. (2015). Industrial Swine and Poultry Production Causes Chronic Nutrient and Fecal Microbial Stream Pollution.?Water, Air, and Soil Pollution,?226(12), 407-420.?? Martin, K. L., Emanuel, R. E., and Vose, J. M. (2018). Terra incognita: the unknown risks to environmental quality posed by the spatial distribution and abundance of concentrated animal feeding operations. Science of the total environment, 642, 887-893. Mayer, P. M., Reynolds, S. K., McCutchen, M. D., and Canfield, T. J. (2007). Meta-analysis of nitrogen removal in riparian buffers. Journal of Environmental Quality, 36(4), 1172–1180. McCulloch, R. B., Stephen Few, G., Murray, G. C., and Aneja, V. P. (1998). Analysis of ammonia, ammonium aerosols and acid gases in the atmosphere at a commercial hog farm in eastern North Carolina, USA. Environmental Pollution, 102(1), 263–268. Michalopoulos, C., Tzamtzis, N., and Liodakis, S. (2014). Effects of an intensive hog farming operation on groundwater in East Mediterranean (I): A study on electrical conductivity, as well as nitrogen and sulfur nutrients. Bulletin of Environmental Contamination and Toxicology, 93(6), 683–687. Miner, J. R., Humenik, F. J., Rice, J. M., Rashash, D. M. C., Williams, C. M., Robarge, W., Harris, D. B., and Sheffield, R. (2003). Evaluation of a Permeable, 5 cm Thick, Polyethylene Foam Lagoon Cover. Transactions of the American Society of Agricultural Engineers, 46(5), 1421–1426. Montefiore, L. R., Nelson, N. G., Dean, A., & Sharara, M. (2022). Reconstructing the historical expansion of industrial swine production from Landsat imagery. Scientific reports, 12(1), 1-12. Mueller, D. K., and Helsel, D. R. (1996). Nutrients in the nation's waters--too much of a good thing? U.S. Geological Survey Circular 1136. National Climatic Data Center. 1994. Asheville, North Carolina. NC DEQ. (2018). North Carolina Nonpoint Source Pollution Management Program. Retrieved January 2, 2022, from https://deq.nc.gov/media/14084/download. 82 NC DEQ. DWR Surface Water Classifications Map. Retrieved November 30, 2022, from https://experience.arcgis.com/experience/7073e9122ab74588b8c48ded34c3df55/page/Page -1/?data_id=dataSource_1- SurfaceWaterClassifications_6584_4677%3A4265&views=Stream-Details NC State Climate Office. Retrieved November 30, 2022, from https://climate.ncsu.edu/ O’Driscoll, M., Johnson, P., and Mallinson, D. (2010). Geological controls and effects of floodplain asymmetry on river–groundwater interactions in the Southeastern Coastal Plain, USA. Hydrogeology Journal, 18(5), 1265–1279. O'Driscoll, M. A. (2012). The 1909 North Carolina drainage act and agricultural drainage effects in eastern North Carolina. Journal of North Carolina Academy of Science, 128(3-4), 59- 73. O’Driscoll, M., DeWalle, D., Humphrey Jr, C., and Iverson, G. (2019). Groundwater Seeps: Portholes to Evaluate Groundwater’s Influence on Stream Water Quality. Journal of Contemporary Water Research & Education, 166(1), 57-78. Osmond, D. L., Hodges, S. H., Hardy, D. H., Lord, W., Kleiss, H. J., Bateman, K., Creamer, N. G., Crozier, C. R., Cubeta, M. A., Jester, W. R., Harrell, N., Green, J., Schultheis, J. R. and Weisz, R. (2003). Tar-Pamlico River Basin Nutrient Management Education. North Carolina Cooperative Extension Service, North Carolina State University. Raleigh, NC. Paerl, H. W., Mallin, M. a., Donahue, C. a, Go, M., and Peierls, B. L. (1995). Nitrogen Loading Sources and Eutrophication. Water Resources Research, 1-119. Panno, S. V., Kelly, W. R., Martinsek, A. T., and Hackley, K. C. (2006). Estimating background and threshold nitrate concentrations using probability graphs. Ground Water, 44(5), 697– 709. Peterjohn, W. T., and Correll, D. L. (1984). Nutrient dynamics in an agricultural watershed: observations on the role of a riparian forest. Ecology, 65(5), 1466-1475. Pinay, G., Ruffinoni, C., Wondzell, S., and Gazelle, F. (1998). Change in groundwater nitrate concentration in a large river floodplain: Denitrification, uptake, or mixing? Journal of the North American Benthological Society, 17(2), 179–189. Poe, A. C., Piehler, M. F., Thompson, S. P., & Paerl, H. W. (2003). Denitrification in a constructed wetland receiving agricultural runoff. Wetlands, 23(4), 817-826. Post, V. E., and von Asmuth, J. R. (2013). Review: Hydraulic head measurements—new technologies, Classic Pitfalls. Hydrogeology Journal, 21(4), 737–750. 83 Puckett, L. J. (2004). Hydrogeologic controls on the transport and fate of nitrate in ground water beneath riparian buffer zones: results from thirteen studies across the United States. Water Science and Technology, 49(3), 47-53. Qiao, C., Liu, L., Hu, S., Compton, J. E., Greaver, T. L., and Li, Q. (2015). How inhibiting nitrification affects nitrogen cycle and reduces environmental impacts of anthropogenic nitrogen input. Global Change Biology, 21(3), 1249–1257. R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Retrieved on September 26, 2022 from https://www.R-project.org/. Ritter, W. F. (1988). Reducing impacts of nonpoint source pollution from agriculture: A Review. Journal of Environmental Science and Health . Part A: Environmental Science and Engineering, 23(7), 645–667. Roberts, T. L. (2007). Right product, right rate, right time and right place… the foundation of best management practices for fertilizer. Fertilizer best management practices, 29, 1-8. Robertson, W. D., and Merkley, L. C. (2009). In?stream bioreactor for agricultural nitrate treatment. Journal of Environmental Quality, 38(1), 230-237. Rosov, K. A., Mallin, M. A., and Cahoon, L. B. (2020). Waste nutrients from U.S. animal feeding operations: Regulations are inconsistent across states and inadequately assess nutrient export risk. Journal of Environmental Management, 269, 110738. Royer, T. V., Tank, J. L., and David, M. B. (2004). Transport and fate of nitrate in headwater agricultural streams in Illinois. Journal of Environmental Quality, 33(4), 1296-1304. Sadeq, M., Moe, C. L., Attarassi, B., Cherkaoui, I., ElAouad, R., and Idrissi, L. (2008). Drinking water nitrate and prevalence of methemoglobinemia among infants and children aged 1–7 years in Moroccan areas. International Journal of Hygiene and Environmental Health, 211(5-6), 546–554. Saunders, D. L., and Kalff, J. (2001). Denitrification rates in the sediments of Lake Memphremagog, Canada–USA. Water Research, 35(8), 1897–1904. Schweigert, P., Pinter, N., and Van der Ploeg, R. R. (2004). Regression analyses of weather effects on the annual concentrations of nitrate in soil and groundwater. Journal of Plant Nutrition and Soil Science, 167(3), 309–318. Shapiro, S. S., and Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3–4), 591–611. Showers, W. J., Eisenstein, D. M., Paerl, H., and Rudek, J. (1990). Stable Isotope Traces of Nitrogen Sources to the Neuse River, North Carolina. Water Resources Research Institute of the University of North Carolina. 84 Sieling, K., and Kage, H. (2006). N balance as an indicator of N leaching in an oilseed rape – winter wheat – winter barley rotation. Agriculture, Ecosystems and Environment, 115(1-4), 261–269. Sigman, D. M., Casciotti, K. L., Andreani, M., Barford, C., Galanter, M., and Böhlke, J. K. (2001). A bacterial method for the nitrogen isotopic analysis of nitrate in seawater and freshwater. Analytical Chemistry, 73(17), 4145–4153. Silva, S. R., Ging, P. B., Lee, R. W., Ebbert, J., Tesoriero, A. J., and Inkpen, E. L. (2002). Forensic applications of nitrogen and oxygen isotopes in tracing nitrate sources in urban environments. Environmental Forensics, 3(2), 125–130. Sloan, A. J., Gilliam, J. W., Parsons, J. E., Mikkelsen, R. L., and Riley, R. C. (1999). Groundwater nitrate depletion in a swine lagoon effluent-irrigated pasture and adjacent riparian zone. Journal of soil and water conservation, 54(4), 651-656. Sousan, S., Iverson, G., Humphrey, C., Lewis, A., Streuber, D., and Richardson, L. (2021). High- frequency assessment of air and water quality at a concentration animal feeding operation during wastewater application to Spray Fields. Environmental Pollution, 288, 117801. Spruill, T. B., Harned, D. A., Ruhl, P. M., Eimers, J. L., McMahon, G., Smith, K. E., and Woodside, M. D. (1998). Water Quality in the Albemarle-Pamlico Drainage Basin; North Carolina and Virginia, 1992–1995; Circular 1157. US Geological Survey: Denver, CO, USA. Spruill, T. B., Showers, W. J., and Howe, S. S. (2002). Application of classification-tree methods to identify nitrate sources in ground water. Journal of Environmental Quality, 31(5), 1538– 1549. Spruill, T. B., Tesoriero, A. J., Mew, H. E., Farrell, K. M., Harden, S. L., Colosimo, A. B., and Kraemer, S. R. (2005). Geochemistry and characteristics of nitrogen transport at a confined animal feeding operation in a coastal plain agricultural watershed, and implications for nutrient loading in the Neuse River Basin, North Carolina, 1999-2002. Scientific Investigations Report. Steenvoorden, J., Fonck, H., and Oosterom, H. P. (1986). Losses of nitrogen from intensive grassland systems by leaching and surface runoff. Nitrogen Fluxes in Intensive Grassland Systems, 85–97. Stewart, A.J. (2001). A simple stream monitoring technique based on measurements of semiconservative properties of water. Environmental Management, 27(1), 37–46. Stone, K.C., Hunt, P.G., Coffey, S.W., and Matheny, T.A. (1995). Water quality status of a USDA water quality demonstration project in the eastern Coastal Plain. Journal of Soil and Water Conservation, 50(5), 567-571. 85 Stone, K.C., Hunt, P.G., Humenik, F.J., and Johnson, M.H. (1998a). Impact of Swine waste application on ground and stream water quality in an eastern coastal plain watershed. Transactions of the ASAE, 41(6), 1665–1670.?? Stone, K. C., Hunt, P. G., Johnson, M. H., and Matheny, T. A. (1998b). Nitrate-n distribution and trends in shallow groundwater on an eastern Coastal Plains watershed. Transactions of the ASAE, 41(1), 59–64. Stone, K. C., Hunt, P. G., Novak, J. M., and Johnson, M. H. (2003). In-stream wetland design for nonpoint source pollution abatement. Applied Engineering in Agriculture, 19(2). Storey, R. G., Williams, D. D., and Fulthorpe, R. R. (2004). Nitrogen processing in the hyporheic zone of a pastoral stream. Biogeochemistry, 69(3), 285–313. Sun, G., McNulty, S. G., Amatya, D. M., Skaggs, R. W., Swift, L. W., Shepard, J. P., and Riekerk, H. (2002). A comparison of the watershed hydrology of coastal forested wetlands and the mountainous uplands in the southern US. Journal of Hydrology, 263(1- 4), 92–104. Thu, K., Donham, K., Ziegenhorn, R., Reynolds, S., Thorne, P. S., Subramanian, P., Whitten, P., and Stookesberry, J. (1997). A control study of the physical and mental health of residents living near a large-scale swine operation. Journal of Agricultural Safety and Health, 3(1), 13. Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R., and Polasky, S. (2002). Agricultural sustainability and intensive production practices. Nature, 418(6898), 671-677.?? Townsend, A. R., Howarth, R. W., Bazzaz, F. A., Booth, M. S., Cleveland, C. C., Collinge, S. K., Dobson, A. P., Epstein, P. R., Holland, E. A., Keeney, D. R., Mallin, M. A., Rogers, C. A., Wayne, P., and Wolfe, A. H. (2003). Human Health Effects Changing Nitrogen Cycle. Frontiers in Ecology and the Environment, 1, 240–246. Twomey, K. M., Stillwell, A. S., and Webber, M. E. (2010). The unintended energy impacts of increased nitrate contamination from biofuels production. Journal of Environmental Monitoring, 12(1), 218-224. UC Davis. (2022). Stable Isotope Facility. University of California at Davis. Retrieved on June 1, 2022 from https://stableisotopefacility.ucdavis.edu/nitrate-no3-water Udy, J. W., Fellows, C. S., Bartkow, M. E., Bunn, S. E., Clapcott, J. E., and Harch, B. D. (2006). Measures of nutrient processes as indicators of stream Ecosystem Health. Hydrobiologia, 572(1), 89–102. US Climate Data. Climate for North Carolina. 2022. Retrieved on April 1, 2022 from https://www.usclimatedata.com/climate/north-carolina/ united-states/3203 86 US EPA. (2000). "Ambient water quality criteria recommendations: information supporting the development of state and tribal nutrient criteria for rivers and streams in nutrient Ecoregion IX.” United States Environmental Protection Agency EPA Report 822-B-00- 019. Retrieved on September 26, 2022 from https://www.epa.gov/sites/production/files/documents/rivers9.pdf US EPA. (2003). National Pollutant Discharge Elimination System permit regulation and effluent limitation guidelines and standards for concentrated animal feeding operations (CAFOs); final rule. Federal Register, 68(29), 7176-7274. US EPA. (2012). NPDES Permit Writers' Manual for CAFOs. US EPA. (2022). “Overview of Greenhouse Gases.” Retrieved December 29, 2022, from https://www.epa.gov/ghgemissions/overview-greenhouse-gases#methane US GS. (2022). The StreamStats Program. Retrieved November 23, 2022 from http://streamstats.usgs.gov USDA/NASS (2021). State Agriculture Overview for North Carolina. Retrieved April 1, 2022, from https://www.nass.usda.gov/Quick_Stats/Ag_Overview/stateOverview.php?state=NORTH +CAROLINA USDA. Web Soil Survey. Retrieved January 1, 2023, from https://websoilsurvey.sc.egov.usda.gov/App/WebSoilSurvey.aspx VanderZaag, A.; Gordon, R.; Jamieson, R.; Burton, D.; Stratton, G. Permeable synthetic covers for controlling emissions from liquid dairy manure. Appl. Eng. Agric. 2010, 26, 287–297. Vandevivere, P., and Baveye, P. (1992). Saturated hydraulic conductivity reduction caused by aerobic bacteria in sand columns. Soil Science Society of America Journal, 56(1), 1–13. Vitousek, P. M., Aber, J. D., Howarth, R. W., Likens, G. E., Matson, P. A., Schindler, D. W., Tilman, D. G. (1997). Human Alteration of The Global Nitrogen Cycle: Sources and Consequences. Ecological Applications, 7(3), 737–7. Vymazal, J. (2017). The use of constructed wetlands for nitrogen removal from agricultural drainage: A Review. Scientia Agriculturae Bohemica, 48(2), 82–91. Walker, J. T., Aneja, V. P., and Dickey, D. A. (2000). Atmospheric Transport and wet deposition of ammonium in North Carolina. Atmospheric Environment, 34(20), 3407–3418. Woli, K. P., David, M. B., Cooke, R. A., McIsaac, G. F., and Mitchell, C. A. (2010). Nitrogen balance in and export from agricultural fields associated with controlled drainage systems and denitrifying bioreactors. Ecological Engineering, 36(11), 1558-1566. 87 Ward, M. H., DeKok, T. M., Levallois, P., Brender, J., Gulis, G., Nolan, B. T., and VanDerslice, J. (2005). Workgroup report: Drinking-water nitrate and health—recent findings and research needs. Environmental Health Perspectives, 113(11), 1607–1614. Wick, K., Heumesser, C., and Schmid, E. (2012). Groundwater nitrate contamination: Factors and indicators. Journal of Environmental Management, 111, 178–186. Wilcoxon, F. (1945) Individual comparisons by ranking methods. Biometrics 1:80–83. Wilhelm, S.R., Schiff, S.L., and Robertson, W.D. 1996. Biogeochemical Evolution of Domestic Waste Water in Septic Systems: 2. Application of Conceptual Model in Sandy Aquifers. Groundwater 34(5): 853-864. Wing, S., Horton, R. A., Marshall, S. W., Thu, K., Tajik, M., Schinasi, L., and Schiffman, S. S. (2008). Air pollution and odor in communities near industrial swine operations. Environmental Health Perspectives, 116(10), 1362–1368. Winner, M. D. and Coble, R. W. (1996). Hydrogeologic framework of the North Carolina Coastal Plain. Professional Paper. Xu, H., Paerl, H. W., Qin, B., Zhu, G., Hall, N. S., and Wu, Y. (2014). Determining critical nutrient thresholds needed to control harmful cyanobacterial blooms in Eutrophic Lake Taihu, China. Environmental Science and Technology, 49(2), 1051–1059. Zeng, Q., Qin, L., Bao, L., Li, Y., and Li, X. (2016). Critical nutrient thresholds needed to control eutrophication and synergistic interactions between phosphorus and different nitrogen sources. Environmental Science and Pollution Research, 23(20), 21008–21019. 88 APPENDIX A: SUMMARY OF LAND COVER A table describing various land cover parameters and the percentage that makes up the farm study area is shown below. Parameter Code Parameter Description Value Unit BASINPERIM Perimeter of the drainage basin as defined in SIR 2004-5262 5.55 miles BSLDEM30FT Mean basin slope, based on slope percent grid 7.03 percent Change in elevation between points 10 and 85 percent of length along main channel to CSL10_85fm basin divide divided by length between points ft per mi 110.27 feet per mi DRNAREA Area that drains to a point on a stream 0.94 square miles ELEV Mean Basin Elevation 260 feet ELEVMAX Maximum basin elevation 393 feet I24H50Y Maximum 24-hour precipitation that occurs on average once in 50 years 7.37 inches LC01BARE Percentage of area barren land, NLCD 2001 category 31 0 percent LC01CRPHAY Percentage of cultivated crops and hay, classes 81 and 82, from NLCD 2001 27.8 percent LC01DEV Percentage of land-use from NLCD 2001 classes 21-24 3.8 percent LC01FOREST Percentage of forest from NLCD 2001 classes 41-43 14.6 percent LC01HERB Percentage of herbaceous upland from NLCD 2001 class 71 50.4 percent LC01IMP Percent imperviousness of basin area 2001 NLCD 0.47 percent LC01SHRUB Percent of area covered by shrubland using 2001 NLCD 0.7 percent LC01WATER Percentage of open water, class 11, from NLCD 2001 1 percent LC01WETLND Percentage of wetlands, classes 90 and 95, from NLCD 2001 1.6 percent LC06BARE Percent of area covered by barren rock using 2006 NLCD 0.1 percent LC06DEV Percentage of land-use from NLCD 2006 classes 21-24 3.8 percent LC06FOREST Percentage of forest from NLCD 2006 classes 41-43 15.2 percent LC06GRASS Percent of area covered by grassland/herbaceous using 2006 NLCD 51.1 percent LC06IMP Percentage of impervious area determined from NLCD 2006 impervious dataset 0.47 percent LC06PLANT Percent of area in cultivation using 2006 NLCD 26.4 percent LC06SHRUB Percent of area covered by shrubland using 2006 NLCD 0.7 percent LC06WATER Percent of open water, class 11, from NLCD 2006 1 percent LC06WETLND Percent of area covered by wetland using 2006 NLCD 1.6 percent LC11BARE Percentage of barren from NLCD 2011 class 31 0.1 percent LC11CRPHAY Percentage of cultivated crops and hay, classes 81 and 82, from NLCD 2011 25.6 percent LC11DEV Percentage of developed (urban) land from NLCD 2011 classes 21-24 3.8 percent LC11FOREST Percentage of forest from NLCD 2011 classes 41-43 36.4 percent LC11GRASS Percent of area covered by grassland/herbaceous using 2011 NLCD 25 percent LC11IMP Average percentage of impervious area determined from NLCD 2011 impervious 0.5 percent LC11SHRUB Percent of area covered by shrubland using 2011 NLCD 6.3 percent LC11WATER Percent of open water, class 11, from NLCD 2011 1 percent LC11WETLND Percentage of wetlands, classes 90 and 95, from NLCD 2011 1.6 percent LC92FOREST Percentage of forest from NLCD 1992 classes 41-43 66.4 percent LFPLENGTH Length of longest flow path 1.561 miles LU92BARE Percent of area covered by barren rock using 1992 NLCD 0 percent LU92DEV Percent of area covered by all densities of developed land using 1992 NLCD 0 percent LU92PLANT Percent of area in cultivation using 1992 NLCD 29.4 percent LU92WATER Percent of area covered by water using 1992 NLCD 1.1 percent LU92WETLN Percent of area covered by wetland using 1992 NLCD 3.1 percent MINBELEV Minimum basin elevation 183 feet OUTLETELEV Elevation of the stream outlet in feet above NAVD88 187 feet PCTREG1 Percentage of drainage area located in Region 1 - Piedmont / Ridge and Valley 0 percent PCTREG2 Percentage of drainage area located in Region 2 - Blue Ridge 0 percent PCTREG3 Percentage of drainage area located in Region 3 - Sandhills 100 percent PCTREG4 Percentage of drainage area located in Region 4 - Coastal Plains 0 percent PCTREG5 Percentage of drainage area located in Region 5 - Lower Tifton Uplands 0 percent PRECIP Mean Annual Precipitation 47 inches PROTECTED Percent of area of protected Federal and State owned land 0 percent SSURGOA Percentage of area of Hydrologic Soil Type A from SSURGO 14 percent SSURGOB Percentage of area of Hydrologic Soil Type B from SSURGO 41.5 percent SSURGOC Percentage of area of Hydrologic Soil Type C from SSURGO 29.1 percent SSURGOD Percentage of area of Hydrologic Soil Type D from SSURGO 14.2 percent Parameter Code Parameter Description Value Unit BASINPERIM Perimeter of the drainage basin as defined in SIR 2004-5262 5.55 miles BSLDEM30FT Mean basin slope, based on slope percent grid 7.03 percent Change in elevation between points 10 and 85 percent of length along main channel to CSL10_85fm basin divide divided by length between points ft per mi 110.27 feet per mi DRNAREA Area that drains to a point on a stream 0.94 square miles ELEV Mean Basin Elevation 260 feet ELEVMAX Maximum basin elevation 393 feet I24H50Y Maximum 24-hour precipitation that occurs on average once in 50 years 7.37 inches LC01BARE Percentage of area barren land, NLCD 2001 category 31 0 percent LC01CRPHAY Percentage of cultivated crops and hay, classes 81 and 82, from NLCD 2001 27.8 percent LC01DEV Percentage of land-use from NLCD 2001 classes 21-24 3.8 percent LC01FOREST Percentage of forest from NLCD 2001 classes 41-43 14.6 percent LC01HERB Percentage of herbaceous upland from NLCD 2001 class 71 50.4 percent LC01IMP Percent imperviousness of basin area 2001 NLCD 0.47 percent LC01SHRUB Percent of area covered by shrubland using 2001 NLCD 0.7 percent LC01WATER Percentage of open water, class 11, from NLCD 2001 1 percent LC01WETLND Percentage of wetlands, classes 90 and 95, from NLCD 2001 1.6 percent LC06BARE Percent of area covered by barren rock using 2006 NLCD 0.1 percent LC06DEV Percentage of land-use from NLCD 2006 classes 21-24 3.8 percent LC06FOREST Percentage of forest from NLCD 2006 classes 41-43 15.2 percent LC06GRASS Percent of area covered by grassland/herbaceous using 2006 NLCD 51.1 percent LC06IMP Percentage of impervious area determined from NLCD 2006 impervious dataset 0.47 percent LC06PLANT Percent of area in cultivation using 2006 NLCD 26.4 percent LC06SHRUB Percent of area covered by shrubland using 2006 NLCD 0.7 percent LC06WATER Percent of open water, class 11, from NLCD 2006 1 percent LC06WETLND Percent of area covered by wetland using 2006 NLCD 1.6 percent LC11BARE Percentage of barren from NLCD 2011 class 31 0.1 percent LC11CRPHAY Percentage of cultivated crops and hay, classes 81 and 82, from NLCD 2011 25.6 percent LC11DEV Percentage of developed (urban) land from NLCD 2011 classes 21-24 3.8 percent LC11FOREST Percentage of forest from NLCD 2011 classes 41-43 36.4 percent LC11GRASS Percent of area covered by grassland/herbaceous using 2011 NLCD 25 percent LC11IMP Average percentage of impervious area determined from NLCD 2011 impervious 0.5 percent LC11SHRUB Percent of area covered by shrubland using 2011 NLCD 6.3 percent LC11WATER Percent of open water, class 11, from NLCD 2011 1 percent LC11WETLND Percentage of wetlands, classes 90 and 95, from NLCD 2011 1.6 percent LC92FOREST Percentage of forest from NLCD 1992 classes 41-43 66.4 percent LFPLENGTH Length of longest flow path 1.561 miles LU92BARE Percent of area covered by barren rock using 1992 NLCD 0 percent LU92DEV Percent of area covered by all densities of developed land using 1992 NLCD 0 percent LU92PLANT Percent of area in cultivation using 1992 NLCD 29.4 percent LU92WATER Percent of area covered by water using 1992 NLCD 1.1 percent LU92WETLN Percent of area covered by wetland using 1992 NLCD 3.1 percent MINBELEV Minimum basin elevation 183 feet OUTLETELEV Elevation of the stream outlet in feet above NAVD88 187 feet PCTREG1 Percentage of drainage area located in Region 1 - Piedmont / Ridge and Valley 0 percent PCTREG2 Percentage of drainage area located in Region 2 - Blue Ridge 0 percent PCTREG3 Percentage of drainage area located in Region 3 - Sandhills 100 percent PCTREG4 Percentage of drainage area located in Region 4 - Coastal Plains 0 percent PCTREG5 Percentage of drainage area located in Region 5 - Lower Tifton Uplands 0 percent PRECIP Mean Annual Precipitation 47 inches PROTECTED Percent of area of protected Federal and State owned land 0 percent SSURGOA Percentage of area of Hydrologic Soil Type A from SSURGO 14 percent SSURGOB Percentage of area of Hydrologic Soil Type B from SSURGO 41.5 percent SSURGOC Percentage of area of Hydrologic Soil Type C from SSURGO 29.1 percent SSURGOD Percentage of area of Hydrologic Soil Type D from SSURGO 14.2 percent 90 APPENDIX B: SOIL SERIES SUMMARY APPENDIX C: SURFACE WATER STATISTICS A statistical summary of the surface water data is shown below. A Kruskal-Wallis was done initially to determine significance within the groups. A post-hoc Wilcoxon test was run to determine which groups showed significance. Significance between data that was normalized was between Upstream (UP) and Downstream (DOW). The rest of the tests were for data from UP, DOW, Seep-Up (SU) and Seep-Down (SD). Parameters were total dissolved nitrogen (TDN), discharge, discharged normalized, TDN exports, TDN exports normalized, and specific conductivity (SC). Kruskal-Wallis TDN Discharge Discharge TDN TDN Exports Conductivity Normalized* Exports Normalized* P Value <0.01 <0.01 0.37 <0.01 <0.01 <0.01 Wilcoxon TDN TDN Exports DO SD SU DO SD SU SD <0.01 SD 0.06 SU <0.01 0.04 SU 0.51 0.63 UP <0.01 <0.01 <0.01 UP <0.01 <0.01 <0.01 Discharge Conductivity DO SD SU DO SD SU SD <0.01 SD 0.01 SU <0.01 1.00 SU <0.01 0.46 UP 0.15 <0.01 <0.01 UP <0.01 <0.01 <0.01 *Data is from UP and DOW only. APPENDIX D: RAW QUALITY AND QUANTITY DATA The following tables contain physiochemical data for surface water and groundwater locations organized by date that were measured in-field and lab-analyzed. Blank locations are due to missed sampling events because of COVID-19 protocol, human and/or lab error, and/or equipment failure. SC=specific conductance (µS/cm); DO=dissolved oxygen (mg/L); Temp=temperature (°C); ORP=oxidation reduction potential (mV); Turb=turbidity (NTU); Q=discharge (L/sec); NQ=normalized discharge (L/ha/sec); DTW=depth to water (cm); NH4=ammonia (mg/L); NO3=nitrate (mg/L); TDN=total dissolved nitrogen (mg/L); Cl=chloride (mg/L) UP 1 (Upstream 1) Date SC DO pH Temp ORP Turb Q NQ NH4 NO3 TDN Cl 27-Nov-19 48 7.66 4.06 12 228 18.1 3.68 0.03 0.48 0.28 0.76 7.26 11-Dec-19 33 5.98 3.6 11 252 5 5.38 0.04 0.04 0.29 0.33 5.79 9-Jan-20 36 10.7 4.23 9.1 244 23.4 22.9 0.19 0.05 0.38 0.43 4.93 17-Feb-20 14 9.3 4.72 11 122 10 15.3 0.13 2.41 0.41 2.82 6.41 9-Mar-20 29 8.6 3.6 10 175 7 11.3 0.09 0.03 0.24 0.54 3.92 1-Apr-20 1-May-20 4-Jun-20 30 6.41 5.05 21 172 30.5 2.94 0.02 0.05 0.44 1.42 4.76 14-Jul-20 30 5.46 4.14 22 267 10.2 2.55 0.02 2.03 0.05 2.27 6.93 31-Aug-20 29 7.74 3.58 23 259 3.5 3.93 0.03 3.09 0.05 3.14 4.20 28-Sep-20 21 6.87 3.55 19 414 30 28.7 0.24 0.03 0.02 0.4 4.26 26-Oct-20 17 6.21 4.43 17 421 9.8 23.1 0.19 0.38 0.32 0.85 5.80 20-Nov-20 26 10.2 3.25 11 463 8.8 25.8 0.21 0.03 0.05 0.41 4.38 15-Dec-20 10 4.17 9 403 10 67.7 0.56 0.1 0.11 0.5 1.97 12-Jan-21 9 8.8 3.98 8.5 397 16.4 27.2 0.22 0.05 0.17 0.49 3.39 94 UP 2 (Upstream 2) Date SC DO pH Temp ORP Turb Q NQ NH4 NO3 TDN Cl- 27-Nov-19 41 8.08 2.78 12 283 0 4.81 0.04 0.03 0.39 0.42 5.4 11-Dec-19 31 4.98 3.1 11 284 3.5 7.93 0.07 0.01 0.09 0.1 4.01 9-Jan-20 30 10.4 3.18 9 289 4.8 27.5 0.23 0.34 0.12 0.46 4.10 17-Feb-20 19 8.9 4.1 11 135 5 18.4 0.15 7.59 0.08 7.67 14.61 9-Mar-20 30 7.9 3.48 10 192 5 7.08 0.06 0.03 0.07 0.34 3.20 1-Apr-20 1-May-20 4-Jun-20 32 6.85 3.66 19 189 6.3 1.33 0.01 0.05 0.03 1.24 3.55 14-Jul-20 30 7.1 3.9 21 271 16 0.63 0.01 0.15 0.04 1.47 3.21 31-Aug-20 29 6.6 3.71 23 239 25 6.76 0.06 0.28 0.04 1.51 2.38 28-Sep-20 24 6.4 3.04 20 416 3.9 22 0.18 0.03 0.1 0.4 3.44 26-Oct-20 17 5.36 2.91 17 496 3.5 22.1 0.19 0.03 0.03 0.42 3.72 20-Nov-20 22 10.4 3.53 11 440 16.5 29.7 0.25 0.02 0.03 0.36 3.31 15-Dec-20 0 3.28 9.5 457 7.9 30.6 0.26 0.03 0.03 0.34 2.49 12-Jan-21 11 11.1 3.45 8.6 429 19.3 42 0.35 0.03 0.06 0.36 3.05 95 DOW (Downstream) Date SC DO pH Temp ORP Turb Q NQ NH4 NO3 TDN Cl- 27-Nov-19 302 9.04 4.9 13 188 20 17.84 0.07 0.15 13.71 13.86 32.66 11-Dec-19 195 5.7 4.83 11 231 17 35.68 0.15 0.04 6.87 6.91 22.51 9-Jan-20 155 12.5 5.05 10 221 12.1 80.14 0.33 0.47 6.73 7.2 15.12 17-Feb-20 95 9.3 6.5 13 128 21 55.22 0.23 0.13 4.58 4.71 11.41 9-Mar-20 144 9.4 5.2 14 101 8 23.79 0.10 0.06 6.59 6.65 14.86 1-Apr-20 1-May-20 4-Jun-20 452 6.3 6.59 22 123 380 7.48 0.03 10.64 30.05 40.69 47.22 14-Jul-20 486 8.3 6.3 24 178 6 2.12 0.01 0.1 25.51 25.61 30.39 31-Aug-20 805 7.65 5.25 25 216 4.6 13.68 0.06 0.1 9.98 10.08 15.24 28-Sep-20 142 7 5.08 21 614 9.4 134.51 0.55 0.05 5.73 5.78 16.07 26-Oct-20 126 7.04 5 19 350 28.5 57.09 0.23 0.2 4.14 4.34 17.47 20-Nov-20 114 10.1 4.65 12 352 12.2 12.85 0.05 0.07 4.06 4.13 11.38 15-Dec-20 124 6.04 10 250 10.3 115.25 0.47 0.13 3.9 4.03 14.31 12-Jan-21 86 9.85 5.23 9.2 312 17.3 141.62 0.58 0.05 2.23 2.28 9.41 96 SU (Seep-Up) Date SC DO pH Temp ORP Turb Q NQ NH4 NO3 TDN Cl- 27-Nov-19 768 9.75 3.77 14.97 220.2 0 3.96 0.08 0.62 58.99 59.61 60.36 11-Dec-19 640 8.65 4.01 13.6 212 0.6 5.38 0.11 0.5 43.37 43.87 46.68 9-Jan-20 750 10.95 4.03 13.81 238.4 0.4 3.96 0.08 0.73 54.39 55.12 55.49 17-Feb-20 697 9.5 4.9 15.34 141 7 2.27 0.04 1.02 48.43 49.45 49.76 9-Mar-20 780 9.29 4.09 14.79 72.6 0 9.06 0.18 0.3 60.61 60.91 60.16 1-Apr-20 1-May-20 4-Jun-20 744 5.66 4.33 26.35 192.2 0.4 4.53 0.09 0.27 53.51 53.78 51.06 14-Jul-20 851 9.12 3.95 26.59 260.5 15 1.39 0.03 0.18 60.1 60.28 61.27 31-Aug-20 781 6.7 4.15 26.3 202 0.5 7.94 0.16 0.76 57.65 58.41 51.99 28-Sep-20 621 5.19 4.07 22.82 409 7 1.98 0.04 0.93 54.3 55.85 55.81 26-Oct-20 356 7.65 4.18 18.83 447.5 0 4.47 0.09 0.69 44.68 45.95 56.14 20-Nov-20 675 7.04 4.18 17.93 422.8 10.6 2.79 0.05 0.67 53.67 54.41 58.70 15-Dec-20 577 5.37 9.47 349 1.6 4.25 0.08 0.36 21.75 22.11 26.17 12-Jan-21 449 8.5 4.72 11.33 342 4.4 1.25 0.02 0.4 30.26 30.66 35.67 97 SD (Seep-Down) Date SC DO pH Temp ORP Turb Q NQ NH4 NO3 TDN Cl- 27-Nov-19 734 9.93 4.04 13.09 219.4 31.8 3.11 0.05 0.37 53.06 53.43 58.16 11-Dec-19 513 8.1 5.8 13.7 197 18.6 1.70 0.03 0.16 37.13 37.29 43.37 9-Jan-20 206 13.15 4.16 11.95 252.3 3.3 10.76 0.17 0.44 14.79 15.23 16.86 17-Feb-20 695 9.9 4.66 18.2 133 0 0.57 0.01 0.18 47.58 47.76 46.05 9-Mar-20 793 9.74 4.26 16.56 108 1 0.85 0.01 0.15 59.91 60.06 59.60 1-Apr-20 1-May-20 4-Jun-20 742 7.5 7.57 28.88 73.8 13.5 16.96 0.27 3.96 13.6 17.56 17.33 14-Jul-20 280 9.28 4.17 27.87 238.5 5.3 0.47 0.01 0.44 19.75 20.19 18.80 31-Aug-20 755 4.83 4.9 29 212 2.9 6.17 0.10 0.36 51.27 51.63 58.03 28-Sep-20 662 4.9 4.3 21.87 439 4.1 0.86 0.01 0.47 53.5 53.97 56.50 26-Oct-20 209 6.28 4.81 18.67 436.7 8.7 0.09 14.77 14.86 20.87 20-Nov-20 655 9.31 4.47 14.61 432.8 0.5 4.49 0.07 0.3 52.63 52.93 56.59 15-Dec-20 690 4.68 8.22 393.2 2.1 2.55 0.04 0.25 30.29 35.47 43.57 12-Jan-21 308 11.8 5.7 10.34 331 32.8 1.88 0.03 0.06 17.69 17.75 23.67 98 Background (BG) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 9-Jan-20 17-Feb-20 9-Mar-20 23 9 5 13.81 174 116.74 0.04 0.08 0.259 3.64 1-Apr-20 1-May-20 4-Jun-20 145 7.6 3.94 16.4 360 130.45 0.08 0.18 1.53 2.92 14-Jul-20 37 204.3 4.96 18.5 40 142.65 0.49 0.09 1.42 2.79 31-Aug-20 31 3.3 4.68 20.9 180 125.88 0.05 0.06 1.33 3.71 28-Sep-20 27 36.5 3.34 20 421.9 104.85 0.03 0.04 0.42 3.88 26-Oct-20 26 5.6 4.85 18.4 142 104.85 0.02 0.03 0.35 4.96 20-Nov-20 31 4.4 5.35 17.2 105 99.97 0.03 0.06 0.43 4.38 15-Dec-20 29 4.78 5.49 14.7 180.9 103.63 0.03 0.06 0.39 5.28 12-Jan-21 21 4.53 6.39 11.9 157 93.27 0.03 0.08 0.40 4.03 99 Lagoon (LAG) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 9640 0.74 7.7 9.99 -293 NA 334.868 0.056 484.72 608.726 9-Jan-20 4657 2.3 8 9.1 -119 NA 502.44 0.06 502.5 803.2501 17-Feb-20 8127 1.26 7.44 15.29 -199.9 NA 499.74 0.04 499.78 618.47 9-Mar-20 5235 0.88 7.78 14.58 -193.5 NA 363.25 0.23 363.48 499.61 1-Apr-20 NA 1-May-20 NA 4-Jun-20 1137 0.48 8.89 29.73 -324.3 NA 678.41 0.39 678.8 903.44 14-Jul-20 18623 0.51 7.49 30.24 -551 NA 1824.46 0.48 1824.94 837.56 31-Aug-20 12544 0.9 8.16 29.45 -280 NA 984.35 0.95 985.3 772.17 28-Sep-20 8584 0 8.44 25 -73.4 NA 26-Oct-20 6975 1.3 8.56 19.7 -190 NA 327.60 0.49 377.10 743.33 20-Nov-20 7305 0.99 8.71 17.55 -267 NA 292.50 0.03 319.20 658.99 15-Dec-20 5484 2.5 8.65 10.67 -176.5 NA 354.66 -0.05 354.61 689.56 12-Jan-21 4142 3.26 9.37 10.36 -261.3 NA 269.82 0.03 304.80 554.25 100 Sprayfield 1 (SF1) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 922 3.2 4.08 13.61 311.1 19.81 9.4 34.06 43.46 215.15 9-Jan-20 892 4.2 4.18 13.64 292 39.32 9.81 27.07 36.88 223.32 17-Feb-20 1189 4.67 4.27 15.08 265.5 31.70 11.24 24.15 35.39 230.03 9-Mar-20 1120 4.7 4.08 13.58 289 49.99 10.19 22.22 32.41 230.27 1-Apr-20 1-May-20 4-Jun-20 1184 4 4.87 21.9 359 78.94 9.53 21.02 30.55 236.85 14-Jul-20 1072 5.06 4.33 25.2 263.3 97.23 6.33 17.67 24 205.73 31-Aug-20 1016 2.8 4.5 26.5 60 81.38 10.13 18.2 28.33 193.79 28-Sep-20 1089 5.11 3.87 23 423.7 32.61 9.6 16.28 26.16 223.77 26-Oct-20 935 3.2 4.2 19.6 217 18.59 9.65 14.9 24.55 229.65 20-Nov-20 962 3.3 4.3 18.1 215 53.64 9.57 12.8 23.1 236.24 15-Dec-20 850 4.25 5.25 14.6 37.8 30.48 9.5 10.26 20.1 239.40 12-Jan-21 801 4.42 6.35 10.6 77 14.33 8 8.49 16.62 245.00 101 Sprayfield 2 (SF2) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 298 6.31 4.28 13.6 345 28.65 0.04 3.32 3.36 20.31 9-Jan-20 225 7.1 4.5 13.6 292 54.56 0.01 2.48 2.49 18.56 17-Feb-20 239 6.98 4.55 14.15 260 14.02 0.06 0.8 1.518 11.3 9-Mar-20 197 6.8 4.5 14.5 256 51.82 0.03 1.72 1.75 8.32 1-Apr-20 1-May-20 4-Jun-20 138 5.67 5.13 21.6 214 77.72 0.02 0.95 1.58 11.22 14-Jul-20 46 4.56 5.66 25.1 100 0.07 0.15 1.25 9.49 31-Aug-20 67 5.05 5.6 25.7 -8 93.57 0.15 0.33 1.51 6.94 28-Sep-20 292 3.83 4.22 22.7 221.1 57.61 0.73 1.17 1.9 16.33 26-Oct-20 272 4.5 4.41 20.2 217.5 46.02 0.02 1.21 1.23 15.90 20-Nov-20 198 7 4.81 18 207 77.42 0.02 0.97 0.99 13.52 15-Dec-20 258 6.05 5.37 14.2 194.6 43.89 0.03 1.78 1.81 23.27 12-Jan-21 265 6.26 6.34 10.9 145.3 30.78 0.02 3.11 3.13 20.25 102 Sprayfield 3 (SF3) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 282 5.36 6.1 13.25 213.1 28.04 0.04 9.89 9.93 22.80 9-Jan-20 236 4.4 6.2 13.5 149 57.61 0.04 3.91 3.95 16.66 17-Feb-20 255 7.23 7.07 13.28 108.8 50.60 0.05 1.22 1.351 8.14 9-Mar-20 221 5.5 6.11 13.35 92.3 63.70 0.13 0.78 1.096 8.26 1-Apr-20 1-May-20 4-Jun-20 370 3.5 5.88 20.5 -18 74.37 0.08 0.23 1.56 5.35 14-Jul-20 276 4.1 6.34 24.1 -60.5 94.18 0.06 0.08 1.6 2.55 31-Aug-20 235 4.2 6.11 25.1 -49 82.60 0.12 0.03 1.77 4.55 28-Sep-20 327 2.24 5.95 22.5 144 41.15 0.17 5.91 6.08 6.49 26-Oct-20 254 3.9 5.94 20.2 220 30.48 0.04 8.66 8.7 5.90 20-Nov-20 298 6.01 6.22 17.1 191.1 86.56 0.06 7.90 9.73 6.46 15-Dec-20 243 7.11 6.81 13.5 88.2 37.80 0.04 14.96 15 4.61 12-Jan-21 208 5.46 7.16 10.5 65.3 21.95 0.01 14.49 14.56 3.04 103 Riparian Buffer 1 (RB1) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 203 4.67 5.12 13.18 79.3 17.53 1.69 5.01 6.7 35.99 9-Jan-20 177 3.7 5.2 12.39 139 21.79 1.56 3.79 5.35 30.71 17-Feb-20 161 3.7 5.13 13.1 158 19.35 1.08 3.09 4.847 25.58 9-Mar-20 146 3 4.9 12.61 112 19.66 1.25 0.22 2.236 20.2 1-Apr-20 1-May-20 4-Jun-20 142 4.9 5.21 18.2 166 44.04 0.87 0.02 2.19 15.99 14-Jul-20 213 2.99 5.45 21.5 21.8 60.50 0.75 10.11 10.86 22.08 31-Aug-20 155 3.62 5.47 22.9 70.8 136.40 0.93 2.23 3.38 12.21 28-Sep-20 148 27.1 5.29 20.7 161 20.57 0.48 0.233315 2.03 16.87 26-Oct-20 157 3.36 5.59 18.9 149 18.14 0.62 2.114903 3.52 15.49 20-Nov-20 137 5 6.22 16.1 -60 23.62 0.71 0.461517 2.32 17.72 15-Dec-20 122 3.57 6.47 13.1 1.7 20.57 0.90 0.209026 2.12 18.19 12-Jan-21 102 3.69 7.32 10.6 -121 15.09 0.86 0.395698 2.12 15.03 104 Riparian Buffer 2 (RB2) Date SC DO pH Temp ORP DTW NH4 NO3 TDN Cl- 11-Dec-19 215 10.85 4.56 13.51 333.8 21.03 0.07 11.09 11.16 44.19 9-Jan-20 209 3.3 4.6 12.5 250 20.42 0.02 11.1 11.12 41.79 17-Feb-20 252 3.46 4.5 13.3 235 18.90 0.11 10.38 10.49 37.14 9-Mar-20 189 4.3 4.9 12.7 151 23.16 0.05 12.04 12.09 38.09 1-Apr-20 1-May-20 4-Jun-20 214 3.7 4.09 17.7 360 39.93 0.01 9.5 9.51 34.01 14-Jul-20 222 2.87 5.09 19.4 203.1 42.37 0.21 8.98 9.19 28.69 31-Aug-20 200 3.8 4.64 22.6 187 28.96 0.01 6.67 6.67 29.32 28-Sep-20 220 0.25 4.49 20.5 289.8 20.73 0.06 4.00 5.1 32.40 26-Oct-20 182 4.4 5.13 18.5 144 17.68 0.06 7.08 7.14 37.00 20-Nov-20 176 4 5.5 16.5 35 18.90 0.03 8.25 8.28 32.26 15-Dec-20 179 2.89 5.4 13.6 172.5 19.51 0.04 8.98 9.01 33.67 12-Jan-21 176 3.64 6.5 11.4 200 17.68 0.02 8.33 8.36 31.85 105 APPENDIX E: PRECIPITATION DATA FROM HIGH FREQUENCY EVENT Daily precipitation data during high frequency event from August 14, 2020 to August 31, 2020. Spray event took place from August 20, 2022 to August 24, 2022. Date Total Precipitation (cm) 14-Aug-20 0.51 15-Aug-20 3.20 16-Aug-20 0.03 17-Aug-20 1.09 18-Aug-20 0.00 19-Aug-20 0.84 20-Aug-20 1.98 21-Aug-20 0.43 22-Aug-20 0.00 23-Aug-20 0.00 24-Aug-20 2.16 25-Aug-20 0.05 26-Aug-20 0.00 27-Aug-20 0.00 28-Aug-20 0.00 29-Aug-20 0.00 30-Aug-20 0.00 31-Aug-20 1.88 APPENDIX F: SPECIFIC CONDUCTIVITY FIELD EVENT Below is a table containing individual specific conductivity (SC) values along the stream and seep and locations where the measurement was taken. SC Longitude Latitude (?S/cm) -78.9466 35.3229 39 -78.9466 35.3231 35 -78.9467 35.3231 23 -78.9465 35.3231 31 -78.9462 35.323 35 -78.9458 35.323 33 -78.9457 35.3232 33 -78.9454 35.3232 33 -78.9453 35.3235 36 -78.945 35.3239 33 -78.9444 35.3241 33 -78.944 35.3246 33 -78.9433 35.3248 35 -78.9434 35.3247 36 -78.9431 35.325 42 -78.9431 35.3236 917 -78.943 35.3249 752 -78.9429 35.325 195 -78.9423 35.3254 130 -78.9423 35.3254 240 -78.9404 35.3263 131 -78.9405 35.3259 587 -78.9404 35.3263 456 -78.9404 35.3263 170 APPENDIX G: SEEP-UP AND SEEP-DOWN WATERSHED ESTIMATE APPENDIX H: GROUNDWATER STATISTICAL SUMMRY A summary of groundwater statistics is show below. A Kruskal-Wallis test was conducted initially to determine significance withing groups. A post-hoc Wilcoxon test was then conducted to determine what groups showed significance. Statistical tests were used to test differences between Background (BG), Lagoon (LAG), Sprayfields (SF, SF1, SF2, SF3), and Riparian Buffers (RB, RB1, RB2). Parameters were total dissolved nitrogen (TDN), ammonium (NH +4 ), nitrate (NO -3 ), temperature (Temp.), and specific conductivity (SC). Temp. was between groundwater and wastewater temperatures from colder months (December-February) and warmer months (June-August). Kruskal-Wallis TDN TDN* NH + -4 NO3 Temp. SC SC* P Value <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 Wilcoxon TDN BG LAG RB1 RB2 Sf1 SF2 LAG <0.01 RB1 <0.01 <0.01 RB2 <0.01 <0.01 0.01 SF1 <0.01 <0.01 <0.01 <0.01 SF2 0.05 <0.01 0.04 <0.01 <0.01 SF3 <0.01 <0.01 1.00 1.00 <0.01 0.81 BG LAG RB LAG <0.01 RB <0.01 <0.01 SF <0.01 <0.01 1.00 NH +4 BG LAG RB1 RB2 Sf1 SF2 LAG <0.01 RB1 <0.01 <0.01 RB2 1.00 <0.01 <0.01 SF1 <0.01 <0.01 <0.01 <0.01 SF2 1.00 <0.01 <0.01 1.00 <0.01 SF3 1.00 <0.01 <0.01 1.00 <0.01 1.00 NO -3 BG LAG RB1 RB2 Sf1 SF2 LAG 1.00 RB1 0.01 0.81 RB2 <0.01 <0.01 <0.01 SF1 <0.01 <0.01 <0.01 0.00 SF2 <0.01 0.01 1.00 <0.01 <0.01 SF3 0.05 0.10 1.00 1.00 <0.01 1.00 110 SC BG LAG RB1 RB2 Sf1 SF2 LAG <0.01 RB1 <0.01 <0.01 RB2 <0.01 <0.01 0.02 SF1 <0.01 <0.01 <0.01 0.00 SF2 <0.01 <0.01 0.94 1.00 <0.01 SF3 <0.01 <0.01 0.00 0.01 <0.01 1.00 BG LAG SF LAG <0.01 SF <0.01 <0.01 RB <0.01 <0.01 <0.01 *Tests were conducted between grouped sites BG, LAG, SF, and RB. 111 APPENDIX I: MIXING MODEL RANGE ESTIMATES Two-component mixing model using minimum (J1) and maximum (J2) chloride and TDN concentrations to estimate TDN mass reductions. Chloride concentrations (Cl-), fraction of wastewater (fraction of WW), fraction of groundwater (fraction of BG), predicted total dissolved nitrogen (predicted TDN), observed total dissolved nitrogen (observed TDN), chloride/total dissolved nitrogen ratios (Cl- /TDN), and total dissolved nitrogen mass reduction (TDN mass reduction) are shown for sites Lagoon (LAG), Sprayfield 1 (SF1), Sprayfield 2 (SF2), Sprayfield 3 (SF3), Riparian Buffer 1 (RB1), Riparian Buffer 2 (RB2), and Background (BG). Table J1. Two component mixing model using minimum chloride and TDN to estimate TDN mass reductions. Observed Predicted TDN Cl Fraction Fraction Site Location TDN Cl/TDN TDN Mass (mg/) of WW of BG (mg/L) (mg/L) Reduction Lag Lagoon 499.6 1.00 0.00 304.8 1.64 SF1 Sprayfield 193.8 0.38 0.62 117.38 16.6 11.67 86% SF2 Sprayfield 6.94 0.01 0.99 2.85 0.86 8.07 70% SF3 Sprayfield 2.55 0.00 1.00 0.16 1.1 2.32 -579% RB1 Buffer 12.2 0.02 0.98 6.08 2.03 6.01 67% RB2 Buffer 29.7 0.05 0.95 16.80 5.1 5.82 70% BG Background 2.71 0.00 1.00 0.26 10.42 Table J2. Two-component mixing model using maximum chloride and TDN to estimate TDN mass reductions. Observed Predicted TDN Cl Fraction Fraction Site Location TDN Cl/TDN TDN Mass (mg/) of WW of BG (mg/L) (mg/L) Reduction Lag Lagoon 903.4 1.00 0.00 1824.9 0.50 SF1 Sprayfield 245 0.27 0.73 488.21 43.5 5.63 91% SF2 Sprayfield 23.3 0.02 0.98 38.11 3.36 6.93 91% SF3 Sprayfield 22.8 0.02 0.98 37.10 14.6 1.56 61% RB1 Buffer 36 0.03 0.97 63.90 10.8 3.33 83% RB2 Buffer 44.2 0.04 0.96 80.55 12.1 3.65 85% BG Background 5.28 0.00 1.00 1.53 3.45