Sediment is the most widespread pollutant of streams and rivers in North Carolina. Construction sites are a source of accelerated erosion contributing to the sediment problem. This study was conducted to determine if the application of polyacrylamide (PAM) to soil surfaces on construction sites reduces erosion and turbidity. Polyacrylamide has been demonstrated to greatly reduce erosion in furrow irrigation, and there is limited evidence it controls erosion when applied to bare soil. Two PAM products applied at manufacturers recommended rates (11.2 and 1.68 kg ha-1) and one half the recommended rates (5.6 and 0.84 kg ha-1) with and without grass seeding and mulching were tested on three North Carolina Department of Transportation construction sites in Raleigh and near Fayetteville. Runoff volumes, turbidity levels, and eroded sediment data were collected after natural rain events. On a 2:1 cut slope, turbidity and sediment loss were significantly decreased with application of seed/mulch. Erosion rates were 20 times greater on bare soil after seven rain events, with or without PAM, compared to treatments receiving seed/mulch. Polyacrylamide applied with seed/mulch produced slight reductions in turbidity and sediment loss in early rain events. At the higher rate, PAM applied directly on a more moderate, 4:1 fill slope decreased sediment loss and turbidity in the first few rain events following application compared to bare soil, with decreases diminishing over time. A sandy fill slope had inconsistent results between PAM treatments but reductions in turbidity and sediment from seed/mulch applications.

Previous field work characterized the wetness requirements of four plant communities by modeling their hydrologic regimes in three reference wetlands and correlating the results. The objectives of this work were to: 1) determine whether the modeled predictions were accurate by growing four tree species that represented different plant communities under the modeled hydrologies in the greenhouse, and 2) determine how different tree species adapt to saturated soil conditions. Bald cypress (Taxodium distichum), sweet bay (Magnolia virginiana), pond pine (Pinus serotina), and swamp chestnut oak (Quercus michauxii) were grown in greenhouse experiments under three hydrologic regimes: i) ponded for 100 d, ii) ponded for 15 d, and iii) unsaturated. Loamy sand and sapric materials from a restored Carolina Bay were the substrate materials used to represent mineral and organic soils. Bald cypress (representing Non-riverine Swamp Forest) adapted well to 100 d of ponding by producing lateral roots near the surface, aerenchyma tissue in roots and stem, and increasing P uptake. In organic soils ponded for 100 d, bald cypress had significantly greater height, diameter, and total biomass than all other tree species. Sweet bay (representing Bay Forest) adapted well to 100 d of ponding by producing adventitious roots on the submerged portion of the stem. Pond pine (representing Pond Pine Woodland) was intolerant to 100 d of ponded conditions and 75% of the seedlings died in the ponded experiments. Swamp chestnut oak (representing Non-riverine Wet Hardwood Forest) was intolerant to ponding and all seedlings died in ponded treatments. A rhizotron study was conducted to further observe and quantify biological, morphological, and physiological changes in two wetland tree species with differing tolerances to saturated conditions. Bald cypress and swamp chestnut oak seedlings were grown under ponded and unponded conditions for 106 d in rhizotrons containing the same soils used in the container study. Within 2 wk of ponding, bald cypress (BC) seedlings in both organic and mineral soils began to produce lateral roots within the top 36 cm of soil. The swamp chestnut oak (SCO) seedlings subjected to ponding did not produce many new roots, and root death was observed within 4 wk of flooding. Scanning electron microscopy images of basal stems and roots of BC subjected to saturation showed substantial development of aerenchyma, which may have aided in the overall tolerance of BC to ponded conditions. This study showed BC seedlings developed numerous adaptations to saturated conditions, which allowed them to grow in wet organic soils of the Coastal Plain. Results from the container and rhizotron experiments showed that modeling of the long-term hydrology of natural communities was accurate in predicting the preferred wetness requirements of representative trees in three of the four plant communities evaluated.

Studies of broiler breeder diet modifications to reduce phosphorus (P) excretion have evaluated bird performance, but no studies have quantified the effects of P in the manure and the impacts to soil and plant availability once soil-applied. Four diets were formulated by factoring two levels, 0.40 and 0.22% available P (NRC, Low, respectively), with or without phytase during the breeder laying phase (wk 22 to 64). Breeders fed phytase produced manures with 15% lower total P concentration, but did not change manure water-soluble P (WSP). However, P in the breeder manures was > 92% orthophosphate. The incubation of the four unique manures in samples of Portsmouth (Typic Umbraquult) and Wagram (Arenic Kandiudult) series generally did not differ in concentrations of Mehlich-3 P, soil WSP, total inorganic P, and total P. Phosphorus-based applications of breeder manures (NRC, Low) and triple superphosphate (TSP, Ca[H2PO4]2 H2O) were applied to a P-deficient, Portsmouth soil in the greenhouse to determine the response of corn (Zea mays). Corn growth was equal among P sources in the initial study, but it tended to be greater in the soils amended with breeder manures in the residual study due to the P applied and the apparent liming effect of the soil. The NRC and Low breeder manures were applied at 39 kg P ha-1 in 2007 at Salisbury (Typic Rhodudult), Lewiston (Aquic Paleudult), and Plymouth (Typic Umbraquult), which ranged in soil P levels. Plymouth included TSP and an untreated control. Corn growth was equal among soils amended with the breeder manures NRC and Low in all site-years and grain P removal was equal five out of six site-years. Grain production, grain P removal, and applied P recovery were equal among P sources in 2007, but the breeder manure treatments were greater than TSP in 2008. Breeder manures should be considered equivalent to TSP in P impacts to the soil and plant availability.

The application of mulch products to disturbed soil is frequently used to decrease soil erosion. The addition of polyacrylamide (PAM) has been demonstrated to reduce erosion even further. We conducted rainfall-simulator and field tests on several types of mulch, hydromulch, and erosion control blankets with varying rates of PAM to determine the relative improvements in erosion control and vegetative establishment achieved by different groundcovers and PAM. We also compared applications of granular (37 and 74 kg ha-1) and aqueous (37 kg ha-1) PAM with straw. The tests were done on 2- x 1-m soil boxes tilted to 32% slope and three field locations with similar slopes. Under simulated rainfall, the addition of 37 kg ha-1 aqueous PAM to any mulch type tended to reduce runoff turbidity, TSS, and sediment loss, with reductions of 50% or more in some cases. In the absence of PAM, increasing hydromulch rate from 1971 kg ha-1 to 2957 kg ha-1 significantly reduced turbidity for an experimental cotton product but not for wood fiber. Granular PAM, when applied at a rate equal to that of aqueous PAM, had a significantly higher mean turbidity and TSS; total sediment loss was reduced by dry-PAM only at the higher rate. The lower rate of granular PAM apparently did not reduce erosion but did flocculate the sediment captured in the runoff collection bucket, reducing turbidity. In field experiments, bonded fiber matrix had significantly lower biomass and vegetative cover than all other treatments at one site but differences were not significant at other sites. In one instance where a cotton fiber matrix test product washed off the slope face, runoff turbidity and TSS was significantly higher than other treatments. There were no instances of straw with PAM significantly improving biomass or vegetative cover compared to straw alone.

Wetland restoration can be difficult due to the time consuming efforts it requires to evaluate the soils and vegetation within the area being considered. The objective of this study was to determine whether apparent soil electrical conductivity (ECa) measured via electromagnetic induction (EMI) (a non-contact method) could be used to identify soil particle size distribution, map units, and selected chemical properties at a Carolina bay wetland restoration site, Juniper Bay. Surveys of ECa were performed at Juniper Bay in December 2000, June 2001, October 2005, and December 2005 using the EM38, GEM-2, and GEM-300 in the horizontal and vertical dipole positions. Cores were dug to depths of 6.1 and 15.2 m based on an equilateral triangle grid and adequate representative sampling of soil map units. Pits were dug to 1- and 1.5-m depths. Core and pit samples were analyzed for particle size distribution and soil chemical properties. Surficial samples were collected from 0- to 15-cm and 15- to 30-cm depths; chemical properties were evaluated on these samples. Correlations of ECa with soil particle size distribution and chemical properties were determined. Linear models were derived relating soil clay content and ECa. Soil ECa was compared with soil organic C and water table depth to determine if either property could be detected using EMI. Also, ECa from different survey dates were compared to determine whether there were short-term (within a day) and long-term (months to years) temporal changes in ECa. The ECa from the initial overall December 2000 survey was strongly correlated with soil core clay content at depths from 0.61 to 1.22 m and 2.74 to 3.96 m. The reason for this is unknown. Initial overall December 2000 ECa was also strongly correlated with pit clay content. No other soil core or pit particle size analysis properties had strong significant and consistent correlations with ECa for the December 2000 survey or for any of the other surveys. Few chemical properties were correlated with ECa; however, ECa was weakly correlated with pit organic C. Results showed that ECa could not be used to delineate soil map unit, but the map unit ECa means were different. However, ECa could not detect water table depth at Juniper Bay. Soil ECa was found to be temporally variable, and ECa surveyors should be conscious of potentially varying conditions from survey to survey. Along with ground-truthing, ECa modeling could be used to estimate clay content at certain depths and locations within Juniper Bay. This could aid wetland restoration by identifying locations of aquitards, which might have been penetrated by drainage ditches. Filling the ditches in the correct places could allow restoration to progress faster and more efficiently.

Recent research indicates that siderophores, a class of biogenic ligands with known exceptional affinity for Fe(III), can also strongly complex Co(III), an element essential to normal metabolic function of microbes and animals. This study was conducted to examine the siderophore-promoted dissolution rates and mechanisms of Co from model synthetic Co-bearing minerals to elucidate the role of siderophores in biogeochemical processes of Co. Dissolution of heterogenite (CoOOH) and four Co-goethites (Co-FeOOH) with different levels of Co substitution were investigated in the presence of a trihydroxamate siderophore, desferrioxamine B (DFOB), using batch and flow-through experiments, respectively. By measuring the complex and total metal concentrations in dissolution products, dissolution rates via multiple pathways were measured as a function of pH. Results showed that DFOB promoted dissolution of Co from Co-bearing minerals via pH-dependent mechanisms. For heterogenite, ligand-promoted dissolution was the dominant pathway at neutral to alkaline pH, while reductive dissolution became dominant for pH < 6. Cobalt substitution in Co-goethite resulted in increased total dissolution rates of both Co and Fe, but ligand-promoted and reductive Co dissolution pathways were difficult to examine due to the slow dissolution rates. The fast dissolution rate of heterogenite, coupled with the high affinity of Co(III) for DFOB, suggests that siderophore-promoted dissolution of Co(III) oxides is a biogeochemically favorable process. Although the association of Co with Fe oxide mineral may limit the Co dissolution rate, siderophore-promoted Co dissolution may still be an effective enough way to increase Co bioavailability. The results also suggest the possibility of radionuclide 60Co by siderophores from recalcitrant Fe oxide phases, which may be important to the fate and transport of 60Co in contaminated environments.

Application of anaerobic swine (Sus scrofa domesticus) lagoon liquid onto cropland by irrigation is a common method of waste disposal and treatment. Currently, the application rate of swine lagoon liquid is based on the N concentration of the lagoon liqu id and the N required by the receiver crop to obtain a realistic yield. In North Carolina, only 50% of the total N in the swine lagoon liquid applied by irrigation is considered available for plant use during the first year after application. Uncertaint y exists as to whether this coefficient accurately predicts the amount of plant-available N. Therefore, research was conducted in the Coastal Plain of North Carolina to determine the efficiency of N uptake by corn (Zea mays L.) and soybean (Glycine max M errill) receiving swine lagoon liquid through irrigation. The line-source sprinkler irrigation method was used to provide a continuous variable N rate, ranging from 0 to 290 kg N/ha, across the field during 1999 and 2000. Ammonia volatilization losses r anged from 6 to 22% during irrigation. Crop yield and grain N recovered were affected more by the amount of liquid than N applied in 1999. Nitrogen recovered in grain in 1999 was <15% for both corn and soybean at 168 kg N/ha of either swine lagoon liqui d or ammonium nitrate. In 2000 at the 168 kg N/ha rate, grain N removal by corn, nonnodulating soybean, and nodulating soybean was 28, 25, and 39% from swine lagoon liquid and 45, 31, and 56% from ammonium nitrate. Based on yields and grain N removed by corn and nonnodulating soybean in 2000, N from applied swine lagoon liquid, accounting for N losses during irrigation, was about 70% as effective as ammonium nitrate. Symbiotic N2 fixation by the soybean was reduced by 60% when applied N reached 175 kg N/ha for both ammonium nitrate and swine lagoon liquid. While nodulating soybean removed more grain N than did either corn or nonnodulating soybean in 2000, soil inorganic N concentrations at the end of the growing season were higher for the nodulating s oybean. Therefore, it is not conclusive if soybean would be a better receiver crop than corn for swine lagoon liquid. Based on the results of this study, using the 50% available N coefficient of the lagoon liquid comes close to predicting plant-availabl e N when N losses during irrigation are around 25%. Nitrogen losses during irrigation can significantly affect plant-available N when applied N is based on the N concentrations of the lagoon liquid.

Antibiotics are used in livestock production for the treatment of diseases and for improvement of feed efficiency and growth. However, agricultural use of antibiotics may be partly responsible for the emergence of antibiotic-resistant organisms. Large amounts of managed manure are land applied, which opens the door for the spread of antibiotic resistance in the environment. Thus, the goal of this project was to evaluate the effects of lagoon treatment on the persistence of antibiotic resistant enteric bacteria isolated from swine feces. Both cool season and warm season samples were collected from a swine farm located in Sampson County, NC. Each season samples included three nursery swine fecal samples, three nursery swine lagoon liquid samples, four finishing swine fecal samples, three finishing swine lagoon liquid samples, and four soil samples from both nursery and finishing swine spray field. A total of 4032 E. coli isolates and 4896 Enterococcus isolates were obtained from the samples. The antibiotic resistance profiles of the isolates were determined using a set of antibiotics at various concentrations. The antibiotic cephalothin, erythromycin, oxytetracycline, tetracycline, streptomycin, and neomycin were tested for both bacterial species, but different concentrations were applied. For E. coli, rifampicin was also tested; for Enterococcus, chlortetracycline, vancomycin, and amoxicillin were also tested. After antibiotic resistance analysis was achieved, 25 isolates were randomly selected from each sample for further evaluation by polymerase chain reaction test. Soil samples were collected; however, fecal indicator bacteria were not recovered. Additionally, E. coli was not recovered from warm season nursery lagoon samples. All isolates displayed multiple antibiotic resistance, and for the isolates from the same source, the resistance patterns were similar for the antibiotics within the same antibiotic family. Percentages of resistant isolates were greater in nursery fecal samples than in finishing fecal samples for majority of antibiotic tests. For nursery samples, percentages of antibiotic resistant isolates decreased after lagoon treatment for majority of antibiotic tests. For finishing samples, no such trend was obvious. The results indicated that antibiotic resistant isolates still persist in the lagoon liquid, which may cause potential risk to human and environmental health. And because antibiotic resistance may affect later therapeutic and subtherapeutic value of these antibiotics, management strategies of agricultural antibiotic use may be improved. The antibiotic resistance patterns and molecular banding patterns of the isolates were not unique to a specific source. The results suggest that there is considerable overlap among nursery feces, nursery lagoon, finishing feces and finishing lagoon samples. However, if combine of the feces and the lagoon isolates together and only classify isolates from nursery to finishing swine, the percentages of correctly classified isolates became larger. The results suggest that ARA and PCR would best be used for identifying fecal contamination from swine sources based on broad categories (nursery versus finishing) instead of relying on these procedures for specific identification of lagoon and feces separately.

The restoration of drained agricultural soils to wetlands may result in the dissolution of P from reduced soils, causing the eutrophication of nearby surface waters. Therefore, continued improvements in the ability to predict P dissolution are necessary to minimize its potential environmental hazard. The objectives of this project were to determine if P dissolution would occur when soil material from a drained Carolina bay wetland was reduced, and to hypothesize explanations of differences in P dissolution between soils based on aqueous solution chemistry. Suspensions (15 g kg-1) of < 53 μm separates from surface samples (0-10 cm) of six poorly drained soils were subjected to microbial reduction for 25 d in a continuously stirred reactor. In a separate experiment, saturated whole soil samples (2.5 g H2O g-1 solids) were incubated under O2-free conditions for 62 d. In addition to total P, Fe, Mn, and Al; dissolved reactive phosphate (DRP), Fe(II), and dissolved organic carbon (DOC) were measured in filtrate samples from both experiments. A net increase in P dissolution (two-fold increase, up to 1.2 mg L-1) was observed for only one of six suspensions (Ponzer 1) in the continuously stirred reactor experiment. In that suspension the molar P:Fe(II) suggested that reductive dissolution of Fe(III)-bound P could not fully account for the P that dissolved, and DOC was highly correlated with P dissolution. For reactor suspensions in which no net P dissolution occurred, oxalate-extractable Al was negatively correlated (p < 0.05) with final [DRP], and DOC concentrations were approximately 2-fold lower than in the Ponzer 1 suspension. In the static incubation experiment, P dissolution occurred in all four samples, and the highest concentration was seen in the Ponzer 1 sample (three-fold increase, up to 2.2 mg DRP L-1). Dissolved organic carbon concentrations were between 2 and 4 fold higher in the static incubation experiment than the highest concentration observed in the stirred reactor experiment, and offer a qualitative explanation for the additional P dissolution that occurred in static incubation experiment. The results of these experiments suggest that P concentrations in soils of the restored wetland will increase upon reduction to levels that are environmentally threatening, and that interaction of DOC with PO4 or minerals that bind PO4 plays an important role in the release of P.

Understanding how management affects long-term site productivity will help sustain forest productivity. In 1991, the USDA Forest Service installed the Long-Term Soil Productivity study in the Croatan National Forest in Craven County, NC. The study objectives were to compare indices of plant-available inorganic N from ion-exchange resin (IER) in bead and membrane form and aerobic incubations in a 8-week study, and to analyze the effects of harvest intensity and site preparation on N availability and tree growth ten years after planting. Site treatments included three levels of organic matter removal and two levels of compaction arranged in a strip-plot design replicated in three blocks. Block 1 consisted primarily of Goldsboro soil, while Blocks 2 and 3 consisted of Lynchburg. The two-month study showed no correlation of the N availability among the three techniques. The short duration of the study and the low soil fertility at the site may have contributed to the lack of correlation, indicating that the two forms of IER may not be equally effective at low nutrient concentrations. Through the 2001 growing season, IER membranes were used to measure soil NO3 and NH4 biweekly. Non-linear analysis indicate that total inorganic N recovered from the IER membranes was significantly influenced by interactions between block, organic matter removal treatment and compaction level (p=0.005). Total basal area of loblolly pines was significantly impacted by block (p=0.015) and organic matter removal treatment (p=0.032). At year ten, treatment effects can still be seen on both soil processes and tree productivity.