H41C-0648
Quantifying runoff water quality characteristics from nurseries and avocado groves subjected to altered irrigation and fertilizer regimes
In agriculture, improper, excessive or poorly timed irrigation and fertilizer applications can result in increased pollutants in runoff and degraded water quality. Specifically, the cultivation of salt sensitive plants and nurseries require significant irrigation and fertilizer that leads to high nutrient leaching. In southern California, a large producer of Avocados and nursery plant, waterways are often subjected to elevated nutrient concentrations, which stress the aquatic ecosystem. In this research, the specific objectives are to determine optimal irrigation and fertilizer application rates for minimizing nutrient and sediment export from avocado groves and nurseries. Altered irrigation and fertilizer application experiments will be implemented and monitored at the San Diego State University's Santa Margarita Ecological Reserve, which contains a 12 ha avocado grove and newly constructed 0.4 ha nursery. The study will last for twelve months, with runoff from natural rainfall or irrigation sampled and analyzed for nutrient concentrations on a monthly basis. The growth rate, leaf nutrient content and plant yield will also be monitored monthly. The nursery site is divided into eight plots (13.5-m x 13.5-m), with each plot containing 1200 plants consisting of four commonly used landscaping varieties in southern California. The avocado grove of the Hass variety is divided into four 1-ha plots. The experimental plots represent combinations of irrigation and fertilization practices with different methods and rates. In all cases, irrigation is fully automated based on soil moisture. To assess the effectiveness of the altered irrigation and fertilizer strategies, runoff water quality and plant yield will be compared to controlled treatments. This research is intended to provide a better understanding of how irrigation and fertilizer management can be used for the long-term reduction of nutrients in the Santa Margarita Watershed, which in turn will lead to improved surface water quality, aquatic habitats, and overall stream health. Preliminary results for runoff water quality (N and P) and plant growth characteristics from two months of monitoring are presented. http://spatialhydro.sdsu.edu
H41C-0649
Lagrangian water quality dynamics in the San Luis Drain, California.
Integration of temporal changes in biological and water quality constituents during downstream transport is critical to understanding aquatic ecosystem and biogeochemical dynamics of rivers, estuaries, and the near- coastal waters into which rivers flow. Changes in chemical, physical, and biological water quality constituents during downstream transport can be evaluated by following a specific parcel of water, known as a Lagrangian study. The objective of this study was to differentiate changes in water quality constituents occurring within a parcel of water as it travels downstream to the changes observed at a fixed sampling location. We sampled a parcel of agricultural drainage water as it traveled downstream for 84 h in a concrete-lined channel (San Luis Drain in San Joaquin Valley) with no additional water inputs or outputs. The Lagrangian sampling occurred in August 2006 and June 2007. Data from the Lagrangian study was compared to data collected at a fixed point using an automatic pump sampler and water quality sonde. Fluorescence (a measure of algal pigments), dissolved oxygen, temperature, pH, and conductivity were measured every 30 minutes, as well as collecting grab samples every 2 h for nutrient and suspended sediment analyses. Sinusoidal diel (24 h) patterns were observed for dissolved oxygen, pH, and temperature within the parcel of water. Algal pigments, nutrients, suspended solids, and turbidity did not exhibit sinusoidal diel patterns, generally observed at a fixed sampling location. The diel patterns observed indicated changes that would occur during downstream transport. Algal pigments showed a rapid day time increase during the first 24 to 48 h followed by a plateau or decrease for the remainder of the study. Algal growth was apparent each day during the study, as measured by increasing dissolved oxygen concentrations, in spite of non-detectable phosphate concentrations (<5 ppb) and nearly complete consumption of soluble silica during the 2007 study. Because current monitoring programs generally entail weekly to monthly (usually midday) sampling, they are insufficient to accurately estimate fluxes of nutrients and organic matter through productive rivers, such as the San Joaquin. Information from this study will assist in developing appropriate monitoring protocols to capture diel variability and also provides process-level information for understanding changes in water quality constituents during downstream transport.
H41C-0650
Unsaturated Zone Nitrate beneath Natural and Agricultural Ecosystems in a Semiarid Region, Southern High Plains, USA
Quantifying nitrate reservoirs in the unsaturated zone is important for linking land surface processes and groundwater contamination. The purpose of this study was to characterize nitrate reservoirs beneath natural ecosystems and rainfed (nonirrigated) agricultural ecosystems using data from an area of high groundwater nitrate contamination (20% of wells > 10 mg/L nitrate-N) in the southern High Plains, Texas, USA as an example. Profiles were drilled beneath natural ecosystems (5), and rainfed (19) agricultural ecosystems in these regions. Previous studies have shown that conversion of land from natural grasslands to cropland increases recharge, displacing chloride bulges that previously accumulated under natural conditions, downward in the profile. Median nitrate-N for profiles beneath natural grassland and shrubland ecosystems is low (0.4 mg/kg dry sediment, range 0.1 – 0.6 mg/kg). Lack of nitrate accumulation under natural ecosystems in the southern High Plains contrasts with large nitrate inventories in the central High Plains and Nevada found in previous studies. Median nitrate-N in the shallow, chloride flushed portion of profiles beneath rainfed agriculture is moderately high (2.6 mg/kg) as a result of fertilizer application. Large nitrate-N inventories (median 10.6 mg/kg) in the transition from low to high chloride concentrations at depth is attributed to nitrate mineralization related to initiation of cultivation. Although nitrogen and oxygen isotopes of nitrate could not distinguish natural from fertilizer nitrate, the chloride profiles and associated age dating of soil pore water can be used to distinguish pre-anthropogenic natural nitrate from fertilizers. Nitrate associated with initial cultivation of soil should pass into the aquifer system as a pulse with a lag time in decades; however, over-application of fertilizers and leaching below the root zone will continue to provide a source of nitrate to the underlying aquifer. Unsaturated zone data are extremely useful in linking surface loading with groundwater nitrate levels and developing a comprehensive understanding of controls and timing of groundwater nitrate contamination.
H41C-0651
Quantifying the capacity of compost buffers for treating agricultural runoff
Agricultural operations, specifically, avocado and commercial nurseries require frequent and significant fertilizing and irrigating which tends to result in excessive nutrient leaching and off-site runoff. The increased runoff contains high concentrations of nutrients which negatively impacts stream water quality. Researcher has demonstrated that best management practices such as compost buffers can be effective for reducing nutrient and sediment concentrations in agricultural runoff. The objective of this research is to evaluate both the hydraulic capacity and the nutrient removal efficiency of: (a) compost buffers and (b) buffers utilizing a combination of vegetation and compost. A series of experiments will be performed in the environmental hydraulics laboratory at San Diego State University. A tilting flume 12-m long, 27-cm wide and 25-cm deep will be used. Discharge is propelled by an axial flow pump powered by a variable speed motor with a maximum capacity of 30 liters per second. The experiments are designed to measure the ratio compost mass per flow rate per linear width. Two different discharges will be measured: (a) treatment discharge (maximum flow rate such that the buffer decreases the incoming nitrogen and phosphorus concentrations below a maximum allowable limit) and (b) breaking discharge (maximum flow rate the buffer can tolerate without structural failure). Experimental results are presented for the hydraulic analysis, and preliminary results are presented for the removal of nitrogen and phosphorus from runoff. The results from this project will be used to develop guidelines for installing compost buffers along the perimeters of nursery sites and avocado groves in southern California. http://spatialhydro.sdsu.edu
H41C-0652
Tracing seasonal nitrate sources and loads in the San Joaquin River using nitrogen and oxygen stable isotopes
The San Joaquin River (SJR) is a heavily impacted river draining a major agricultural basin in central California. This river receives nitrate inputs from multiple point and non-point sources including agriculture, livestock, waste water treatment plants, septic systems, urban run-off, and natural soil leaching. Nitrate inputs to the SJR may play a significant role in driving algal blooms and reducing overall water quality. The San Joaquin River discharges into the San Francisco Bay-Delta ecosystem, and reduced water quality and large algal blooms in the SJR may play a significant role in driving critically low oxygen levels in the Stockton Deep Water Shipping Channel. Correct identification of the major nitrate sources to the SJR is important for coordinating mitigation efforts throughout the SJR-Delta-San Francisco Bay region. Measurements of the nitrogen and oxygen isotopic composition of nitrate were made monthly to bimonthly from 2005 through 2007 within the Lower SJR, major tributaries, and various other water input sources in order to assess spatial and temporal variations in nitrate inputs and cycling in this heavily impacted watershed. The oxygen and hydrogen isotopic composition of water was also measured to better distinguish water sources and identify changes in water inputs. A very wide range of δ15N-NO3 and δ18O-NO3 values were observed in the main stem SJR and tributaries. The δ15N values ranged from +2 to +17 ‰, and the δ18O values ranged from -1 to +18 ‰. Except for a major agricultural drain site (San Luis Drain), all the sites showed temporal changes in both δ15N-NO3 and δ18O-NO3 much greater than the differences seen between individual sites. In general, the δ15N values of nitrate in the larger tributary rivers (Merced, Tuolumne and Stanislaus) were much lower than those of the main stem SJR from April to May; however, after June the tributary values began to rise toward the values in the main stem river. Some of the highest δ15N-NO3 values observed occurred in the Merced River during the latter half of the year. The general increase in δ15N with nitrate concentration, both downstream and during the low flow period, is consistent with increasing amounts of nitrate derived from waste in the downstream section of the SJR and increased agricultural inputs during the summer. Additionally, the influence of denitrification on the δ15N-NO3 values in the SJR is still under investigation.
H41C-0653
Occurrence of Antibiotics and Emerging Contaminants in Dairy Farming
Intense animal husbandry is of growing concern as a potential contamination source of a variety of emerging contaminants including pathogens, naturally occurring and synthetic steroid hormones, and various pharmaceuticals, particularly antibiotics. For example, more than twenty million pounds of antibiotics are sold for use in animal husbandry with 95% going towards therapeutic use. Here, we focus on the application and potential environmental occurrence of pharmaceuticals and disinfectants on dairies. Recommended drug applications are available from national databases. Statistical data on actual usage, however, are not available. We complement national data with interviews and dairy visits for further evaluation of drug and chemical usage (not including pesticides used on crops and fertilizer) and an overall assessment of the potential antibiotics output in dairy waste. We find that aminoglycosides, tetracyclines, and coccidiostats make up much of the total mass of antibiotics used. On dairies using the ionophoric antibiotic monensin as feed additive, monensin makes up a large fraction of the total antibiotics use (by mass). Other chemicals of potential concern include disinfectants used to prevent mastitis, detergents used in the milking parlor, footbath reagents to prevent and treat lameness, and insecticides used to control flies and mites. http://groundwater.ucdavis.edu
H41C-0654
Improving Managed Aquifer Recharge Operation to Reduce Nutrient Load in an Agricultural Basin: Delineation of Processes, Controls, and In-situ Potential
Nitrate is the most common nonpoint source pollutant in surface and ground water in the United States, and is a problem particularly in basins developed for agriculture. There is growing municipal and environmental demand for fresh water in basins that have been influenced by decades of agricultural activity. The goal of this research is to assess the potential for a managed aquifer recharge (MAR) system to improve water quality, with an emphasis on reducing the nitrate load to underlying aquifers. The Pajaro Valley Water Management Agency (PVWMA), in central coastal California, currently operates a MAR project that is permitted to divert and recharge up to 2.5 x 106 m3/yr (2000 ac-ft/year) from a slough (wetland) to augment available ground water supplies. As a result of agricultural runoff and infiltration, diverted slough water is often rich in nitrate, as is the water in the underlying aquifer. However, nitrate concentrations in water samples recovered from the aquifer soon after MAR percolation are often relatively low, suggesting that nitrate may be removed as water percolates from the pond into the aquifer. Autonomous Osmosampler systems were deployed in the recharge pond and four nearby monitoring wells, as part of a pilot study, to collect fluid samples during and after pond operation. Samples collected with these instruments recorded the chemical arrival of water in the aquifer soon after percolation began, in some cases showing a 50% reduction in the concentration of nitrate. The chemical response in the aquifer recorded by the Osmosamplers was consistent with pressure data collected simultaneously in the monitoring wells, demonstrating that Osmosamplers should be useful tools for investigating changes in water quality associated with MAR operation. As this research project becomes fully developed during the 2007-08 water year, we will install Osmosampler systems in ground water monitoring wells surrounding the pond, and will collect shallow fluid samples using piezometers and lysimeters in the base of the pond, to asses chemical transformations that take place during MAR operation. We are particularly interested in quantifying the extent, rates and controls on denitrification, where and when nitrate loads are reduced, and under what physical and chemical conditions. Results of this study should have broad implications for the operation of MAR systems where available water has high nutrient concentrations.
H41C-0655
Denitrification of Shallow Groundwater in an Agriculture Area, Korea
This study was initiated to assess the potentials for nitrate contamination of water supply from the up-gradient agricultural areas. The study site is a small farmland, located in the upper stream of Gwang-gyo reservoir, a water supply source for the Suwon City, Korea. Livestock manure has been used in many of small farmlands as a natural fertilizer in the area. Water levels and quality has been monitored 12 monitoring wells installed into the shallow unconfined aquifer from a since Feb., 2007. Field survey with sampling was conducted in February, April and June representing before, in the beginning of and after the application of fertilizer, respectively. Nitrate levels in groundwater exceeded the Korean Drinking Water Guideline of 10 mg/L of NO3-N only in April samples of mw12 and mw7 with 10.2 mg/L and 11.2 mg/L, respectively. During that sampling period, much manure was found on the land surface, indicating fertilizer application for farming. However, mw8, only 7-m apart from mw7, showed nitrate level of 1.5 mg/L NO3-N. The mw8 shows lower DO and higher DOC, HCO3, and Fe and Mn concentrations than those of mw7, implying more reduced conditions. Consequently, the significant spatial variation in nitrate levels was attributed to the biogeochemical reactions and subsequent denitrification in the shallow aquifer.
H41C-0656
Current-Use Pesticides and Organochlorine Compounds in Precipitation and Lake Sediment from Two High-Elevation National Parks
Current-use pesticides (CUPs) and banned organochlorine compounds (OCCs) were measured in precipitation and lake sediments from two national parks in the Western U.S. to determine their occurrence and distribution in high-elevation environments. CUPs frequently detected in snow were endosulfan, dacthal, and chlorothalonil in concentrations ranging from 0.07 to 2.4 ng/L. Of the OCCs, only chlordane, HCB, and two PCB congeners were detected. Pesticides most frequently detected in rain were atrazine, carbaryl, and dacthal in concentrations from 3.0 to 95 ng/L. Estimated annual deposition rates in one of the parks were 8.4 μg/m2 for atrazine, 9.9 μg/m2 for carbaryl, and 2.6 μg/m2 for dacthal of which over 85% occurred during summer. DDE and DDD were the most frequently detected OCCs in lake sediments. DDD and DDE concentrations in an age dated sediment core suggest that atmospheric deposition of banned OCCs, to high-elevation areas have been in decline since the 1970s. Dacthal and endosulfan sulfate were present in low concentrations (0.11–1.2 μg/kg) and were the only CUPs detected in lake sediments.
H41C-0657
Fate and Transport of Acrylamide in Soil and Groundwater Systems: Sorption, Retardation and Numerical Simulations
Acrylamide (AMD) is a known animal and suspected human carcinogen and is used to produce polyacrylamide (PAM), which has been proposed as a technology for seepage control in unlined water delivery canals. The fate and transport of AMD in groundwater systems is not well known, and previous studies have not quantified the sorption coefficient (Kd), sorption isotherms, or estimated AMD breakthrough and transport parameters in soil and water systems. In this study, batch experiments and repacked soil column tests were conducted on three soils (a control sand, and field collected sand and loam soils) to determine the Kd, retardation factor and the form of the sorption isotherm. A numerical model (HYDRUS 2D) was used to simulate a canal environment using the fate and transport parameters of AMD obtained in the laboratory. Microbial degradation rates, obtained from a companion study (Labahn et al. 2007) were used in the model. Photodegradation rates for AMD were also considered. Results from batch studies indicate a Freundlich-type sorption isotherm for AMD in the loam soil. Sorption in the sands was not significant. The preliminary results for the soil column tests show that AMD is conservative in all three types of soil tested with retardation (R) values ranging from 0.985-1.072, with most column studies showing 0.99<R<1.03. AMD recovery rates ranged from 81-105%, with most rates found between 93-100%. The numerical simulation of AMD transport from the canal to underlying groundwater showed significant impact of canal sealing and biodegradation, but little impact from the modest retardation rates and sorption obtained from the laboratory studies.
H41C-0658
Baseflow Indicators for Groundwater Protection Regions
With shifting climatic patterns that affect precipitation volume and frequency, compounded with increasing demands for water, better understanding of the groundwater links to surface water recharge are increasingly in demand. The objective of our research is to provide metrics to determine regions requiring groundwater protection measures based on baseflow predictions. In 2005, a moderate drought occurred within the Fox-Wolf Basin in Northeast Wisconsin. We took 176 discharge spot measurements in the headwater regions of the basin during the summer of 2005 and performed 10-12 repeat measurements at 46 sites during the summer of 2005 and 2006, for a total of 306, measurements. Our confirmed hypothesis is that stream length (distance from headwater to spot measurement) correlates better with baseflow discharge than watershed area. Additionally, we found that landuse had a subtle effect on metrics. Finally, our measurements were compared with USGS spot discharge measurements taken during a severe drought in 1970s. The two data sets show a high degree of correlation but indicate that baseflow water levels are significantly and systematically lower.
H41C-0659
Evaluating the Effectiveness of Native Grass Riparian Buffer Strips to Reduce Pesticide Runoff
Organophosphate pesticides such as diazinon have been a major source of non-point source water pollution in the Sacramento Valley watershed of central California. Diazinon is commonly listed as a pollutant for many tributaries of the Sacramento River on the US Clean Water Act section 303(d) list of impaired waterways. This pesticide is applied either aerially or as a foliar spray to nut and stone-fruit orchards during dormancy, which coincides with the rainy season in northern California. A study was conducted to determine if planting native grasses in the riparian zone was effective in reducing the amount of diazinon entering the surface water in streams flowing through these orchards. Native grasses have deeper root systems and were hypothesized to be more effective in sorbing diazinon and preventing its runoff than non-native grasses. In 2004, nine 20 foot by 20 foot riparian buffer plots were constructed along the banks of the South Fork of Walker Creek, west of the town of Orland in the Sacramento Valley. Three of the nine plots were maintained as bare ground, three were left with resident weeds including dense non-native grasses, and three were planted with native grasses, which included purple needlegrass (Nassella pulchra), creeping wildrye (Elymus triticoides), and deergrass (Muhlenbergia rigens). The experimental design simulated orchard runoff by applying mixtures of water and diazinon at observed field concentrations. The pesticide load was evenly applied across the top of each buffer plot at a rate consistent with local runoff rates in an average storm. Rainfall on the buffer plots was simulated with overhead sprinklers at a rate of 0.75 inches per hour, also an average storm for this area. Runoff was monitored at the downslope side of the plots with flumes funneled to large holding tanks. From these tanks, composite water samples were collected after runoff had ceased. The samples were analyzed for diazinon concentration, nitrates, and total suspended sediment. Results were similar to a demonstration and reconnaissance project conducted in 2003, which concluded that vegetative plots were more effective than the bare ground control in removing pesticide. However, the native grasses were not significantly more effective in reducing the diazinon concentration than non-native grasses. Vegetated buffer strips can help reduce pesticide loadings from orchards to local surface water systems.
H41C-0660
Phosphorus fluxes in headwater streams draining non-research poultry-pasture operations in north-central Georgia, USA
Poultry production is the largest agricultural commodity in Georgia, USA. Due to inefficient utilization of the phosphorus (P) in poultry feed, the manure contains high concentrations of P. When used as fertilizer for crops and pasture, poultry manure may be washed from the soil surface and increase eutrophication risks to downstream lakes and reservoirs. Long term application of poultry manure may result in P saturation of the soils. In the upper Etowah River basin in north-central Georgia, a long history of poultry farming has resulted in high P levels in soils receiving regular poultry manure applications. Few studies to date have been performed on the estimation of P fluxes from operational commercial poultry farms in Georgia. In Fall 2006, a 20-month surface water quality monitoring program was completed that was aimed at estimating P and suspended sediment fluxes in nine headwater streams draining poultry-pasture operations in the upper Etowah River basin. The nine catchments differed in terms of land use history, soil P levels, best management practices and other factors. An additional three streams draining U.S. National Forest were also monitored to provide reference concentrations and loads. Monitoring data included continuous (5-minute) streamflow, rainfall, and water quality samples. Water quality samples included biweekly grab samples plus storm samples collected using conventional autosamplers. Storm sampling using autosamplers included collection of discrete samples and composite samples. In particular instances, the two types of storm sample were collected simultaneously. Discrete storm sampling methods enabled collection of both rising and falling hydrograph limb samples to identify potential hysteretic water quality effects. Water samples were analyzed for total P, filterable reactive P, and total suspended solids. We are using this data to compare different flux estimation methods with emphasis on regression models that utilize laboratory results of both discrete and composite samples as well as models using non- transformed data. Preliminary results demonstrate some success in the use of regression models from agricultural streams where P levels in stormflow are high. This may be partially attributed to the combination of the temporal resolution of the monitoring data plus the accommodation of sampling different flow regimes. In streams where concentrations and overall variability of P is low, averaging methods may be the most appropriate.
H41C-0661
Tetracycline Resistance in the Subsurface of a Poultry Farm: Influence of Poultry Wastes
Concentrated animal feeding operations (CAFOs) are considered to be important man-made reservoir of antibiotic resistant bacteria. Using the electromagnetic induction (EMI) method of geophysical characterization, we measured the apparent subsurface electrical conductivity (ECa) at a CAFO site in order to assess the movement of pollutants associated with animal waste. The map of ECa and other available data suggest that (1) soil surrounding a poultry litter storage shed is contaminated by poultry waste, (2) a contamination plume in the subsurface emanates from that shed, and (3) the development of that plume is due to groundwater flow. We focused on understanding the spread of tetracycline resistance (Tc\mbox{\tiny R}), because tetracycline is one of the most frequently used antibiotics in food animal production and therefore probably used at our field site. Microbiological experiments show the presence of Tc\mbox{\tiny R} bacteria in the subsurface and indicate higher concentrations in the top soil than in the aquifer. Environmental DNA was extracted to identify CAFO- associated Tc\mbox{\tiny R} genes and to explore a link between the presence of Tc\mbox{\tiny R} and CAFO practices. A "shot-gun" cloning approach is under development to target the most prevalent Tc\mbox{\tiny R} gene. This gene will be monitored in future experiments, in which we will study the transmission of Tc\mbox{\tiny R} to naive E.~coli under selective pressure of Tc. Experimental results will be used to develop a mathematical/numerical model in order to describe the transmission process and to subsequently make estimates regarding the large-scale spread of antibiotic resistance.
H41C-0662
Assessing the effects of transient and long term phosphorous storage on the total phosphorous yields in distributed hydrologic model
On average most phosphorus (P) entering streams from agricultural fields arrives in particulate form (PP), and so a key to understanding long-term dynamics of P delivery to downstream water bodies is to use distributed hydrologic and sediment transport models. Most such models are keyed off of digital elevation models, and these underpredict P dynamics with respect to its deposition within stream channels. We showed that reservoirs could be used as an appropriate proxy for in-stream pools and suitable for quantifying in-channel sediment and PP storage in Dorn Creek, within the Mendota Watershed in southern Wisconsin. Our data suggest that the PP gives up soluble P (SP) during inter-storm periods, while PP is delivered during storms, and so actual P delivery downstream can be significantly delayed or modified en route. This has potential impact on downstream algae and other P consumers, and consequently is important for connected agricultural activities to water quality. In the nearly 10 km length Dorn Creek, total 47 points were sampled and analyzed to represent all stream conditions. Simulated and measured sediment and P storage were compared by placing reservoirs along the main channel sub-basins in the Agricultural Policy Environmental eXtender (APEX) model. We used the spatially distributed PP storage dynamics from APEX to drive a second model to release SP from in-stream storage and explain observed inter-storm SP discharge and long-term PP retention within the streams. The results support a need for improved model logic that deals with these complex PP-SP delivery processes.
H41C-0663
Isotopic Evidence of Nitrate Sources and its Relationship to Algae in the San Joaquin River, California
Many competing demands have been placed on the San Joaquin River including deep water shipping, use as agricultural and drinking water, transport of agricultural and urban runoff, and recreation. These long-established demands limit the management options and increase the importance of understanding the river dynamics. The relationships among sources of water, nitrate, and algae in the San Joaquin River must be understood before management decisions can be made to optimize aquatic health. Isotopic analyses of water samples collected along the San Joaquin River in 2005-2007 have proven useful in assessing these relationships: sources of nitrate, the productivity of the San Joaquin River, and the relationship between nitrate and algae in the river. The San Joaquin River receives water locally from wetlands and agricultural return flow, and from three relatively large tributaries whose headwaters are in the Sierra Nevada. The lowest nitrate concentrations occur during periods of high flow when the proportion of water from the Sierra Nevada is relatively large, reflecting the effect of dilution from the big tributaries and indicating that a large fraction of the nitrate is of local origin. Nitrogen isotopes of nitrate in the San Joaquin River are relatively high (averaging about 12 per mil), suggesting a significant source from animal waste or sewage and/or the effects of denitrification. The d15N of nitrate varies inversely with concentration, indicating that these high isotopic values are also a local product. The d15N values of nitrate from most of the local tributaries is lower than that in the San Joaquin suggesting that nitrate from these tributaries does not account for a significant fraction of nitrate in the river. The source of the non-tributary nitrate must be either small unmeasured surface inputs or groundwater. To investigate whether groundwater might be a significant source of nitrate to the San Joaquin River, groundwater samples are being collected monthly from over 20 bank and in-stream wells. Preliminary data suggest that much of the groundwater nitrate has been variably denitrified thereby increasing its d15N values, but not by enough to account for the high d15N values in the river nitrate. The d15N of algae in the San Joaquin reflects the high values of the nitrate in the river indicating (1) that the San Joaquin is productive despite its relatively high opacity, (2) that the algae use the nitrate as a primary nutrient source, and (3) that the concentrations of algae in the San Joaquin are not principally dependent on algae from the tributaries being flushed into the river as has been suggested. The sources of nitrate to the San Joaquin River must be identified if algae production is to be controlled and hypoxic conditions in the downstream reaches eliminated.
H41C-0664
Fate of Acrylamide in Soil and Groundwater Systems: Microbial Degradation
Acrylamide monomer (AMD), a suspected human neurotoxin and carcinogen, is present as a contaminant (up to 0.05%) in commercial preparations of polyacrylamide (PAM). PAM is currently being evaluated for wide-spread use as a temporary water-delivery canal sealant across the western United States. To better constrain potential risks associated with PAM applications, we examined the capacity of natural canal microorganisms to degrade AMD in laboratory and field experiments. Dilution cultivation and enrichment approaches were employed to determine the abundance of culturable microorganisms in several canal habitats which can utilize AMD as a sole nitrogen source (typically 104-106/mL) and a collection of isolates was developed. AMD-degrading microorganisms in our collection fell within a limited diversity of genera including Arthrobacter, Xanthomonas, and Pseudomonas; with the latter demonstrating highest capacity for degrading AMD under laboratory conditions. One strain of Pseudomonas fluorescens, isolated from Klamath Irrigation District (Klamath Falls, OR) canal sediment, was chosen for further study in part because this species is well-studied and ubiquitous. The potential for microbial AMD degradation was tested under laboratory conditions using this strain in repacked short (15 cm) column tests with two relevant soil types (sand and loam). Subsequently, the capacity of mixed natural microbial populations to degrade AMD was examined using soil cores collected from the Highline Canal (Rocky Ford, CO), and canal water/sediment slurries with spiked (5 ppm AMD) in situ bottle tests. Degradation of the monomer in the repacked column experiments was evaluated using a step input of 5 ppm AMD and the canal columns were tested with a range of AMD concentrations (1-5 ppm) followed by quantification with an HPLC. The repacked soil columns inoculated with P. fluorescens demonstrated 80-100% AMD degradation within 12 hours. Natural microbial communities in fresh canal sediment columns produced varying levels of AMD degradation, ranging from 40-50% after 36 hours. The in situ bottle test resulted in 50% degradation after 72 hours. Experimental AMD degradation rates and transport parameters, such as sorption and retardation, were combined to model AMD transport in canal and ground water systems (Arrowood et al. 2007). Initial calculations demonstrate the importance of the microbial role in removal of the monomer and infer possible implications informing the development of safe and effective PAM application protocols.
H41C-0665
Dissolved Organic Carbon as a Drinking Water Constituent of Concern in California Agricultural Watersheds
Dissolved organic carbon (DOC) from the breakdown of plant and animal material is a concern for drinking water quality in California due to the potential formation of carcinogenic byproducts during disinfection. Agricultural DOC loading to surface water is a significant concern, but the sources and reactivity in agricultural runoff remains poorly understood. Here we present data on DOC dynamics in surface water from the Willow Slough watershed, a 425\- km2 agricultural catchment in the Sacramento Valley, California. Samples collected weekly during 2006 and 2007 were analyzed for DOC concentration, optical properties (UV absorbance and fluorescence), 13C\- DOC isotopes, and trihalomethane formation potential (a regulated disinfection byproduct formed during chlorination). DOC concentrations at the watershed mouth ranged from 2 to 4 mg/L during winter and spring, with a clear increase in DOC concentrations to more than 7 mg\/ L following the onset of summer irrigation. The 13C\- DOC values revealed a large range (-19 to -27 ‰), with lowest values during winter baseflow and higher values during summer and winter storms. Spectral slopes also varied seasonally (0.012 to 0.020), with steeper slopes during winter baseflow. Both isotopic and optical data provide evidence for algal\- derived DOC during the winter baseflow and terrestrial sources during winter storms and summer irrigation. Total THM formation potential was higher in winter than summer, and is strongly correlated to DOC concentrations in surface waters (r2 = 0.87). In contrast to the total THM formation potential, the specific THM formation potential (e.g., total THM normalized to DOC) decreased during the summer irrigation season, suggesting a change in reactivity related to DOC source or degradation. Additional data from plant leachates and ground water will be discussed, as well as the implications of watershed management on DOC dynamics and reactivity in agriculturally-dominated landscapes.