Clint C. Truman

SURFACE RUNOFF AND LATERAL SUBSURFACE FLOW AS A RESPONSE TO CONSERVATION TILLAGE AND SOIL-WATER CONDITIONS

Conservation tillage has significant potential as a water management tool for cotton production on sandy, drought-prone soils. Plant residue remaining at the soil surface from prior crops serves as a vapor barrier against water loss, reduces raindrop impact energy, slows surface runoff, and often increases infiltration. By increasing infiltration, the potential for greater plant-available water can be enhanced and irrigation requirements reduced. Five years of data were collected to quantify the hydrologic differences between strip till and conventional till production systems. Surface runoff and lateral subsurface flow were measured on six 0.2 ha plots in South Georgia in order to quantify the water-related effects of conservation tillage. Significant differences in surface and subsurface water losses were observed between the conventional and strip tilled plots. Surface runoff from the conventionally tilled plots exceeded that from the strip tilled plots, while subsurface losses were reversed. Surface runoff losses from the conventionally tilled plots exceeded those from the strip tilled plots by 81% (129 mm/year). Shallow lateral subsurface losses from the strip tilled plots exceeded those from the conventionally tilled plots by 73% (69 mm/year). Overall, a net annual gain of 60 mm of water was observed for the strip tilled plots.

Rainfall and tillage effects on transport of fecal bacteria and sex hormones 17β-estradiol and testosterone from broiler litter applications to a Georgia Piedmont Ultisol

Poultry litter provides nutrients for crop and pasture production; however, it also contains fecal bacteria, sex hormones (17beta-estradiol and testosterone) and antibiotic residues that may contaminate surface waters. Our objective was to quantify transport of fecal bacteria, estradiol, testosterone and antibiotic residues from a Cecil sandy loam managed since 1991 under no-till (NT) and conventional tillage (CT) to which either poultry litter (PL) or conventional fertilizer (CF) was applied based on the nitrogen needs of corn (Zea mays L) in the Southern Piedmont of NE Georgia. Simulated rainfall was applied for 60 min to 2 by 3-m field plots at a constant rate in 2004 and variable rate in 2005. Runoff was continuously measured and subsamples taken for determining flow-weighted concentrations of fecal bacteria, hormones, and antibiotic residues. Neither Salmonella, nor Campylobacter, nor antimicrobial residues were detected in litter, soil, or runoff. Differences in soil concentrations of fecal bacteria before and after rainfall simulations were observed only for Escherichia coli in the constant rainfall intensity experiment. Differences in flow-weighted concentrations were observed only for testosterone in both constant and variable intensity rainfall experiments, and were greatest for treatments that received poultry litter. Total loads of E. coli and fecal enterococci, were largest for both tillage treatments receiving poultry litter for the variable rainfall intensity. Load of testosterone was greatest for no-till plots receiving poultry litter under variable rainfall intensity. Poultry litter application rates commensurate for corn appeared to enhance only soil concentrations of E. coli, and runoff concentrations of testosterone above background levels.

Herbicide incorporation by irrigation and tillage impact on runoff loss.

Runoff from farm fields is a common source of herbicide residues in surface waters. Incorporation by irrigation has the potential to reduce herbicide runoff risks. To assess impacts, rainfall was simulated on plots located in a peanut (Arachis hypogaea L.) field in Georgias Atlantic Coastal Plain region after pre-emergence application of metolachlor (2-chloro-N-(2-ethyl-6-methylphenyl)-N-[(1S)-2-methoxy-1-methylethyl]-acetamide) and pendimethalin (N-(1-ethylpropyl)-3,4-dimethyl-2,6-dinitro-benzenamine). Runoff, sediment, and herbicide loss as function of strip tillage (ST) versus conventional tillage (CT) were compared with and without irrigation (12.5 mm) after application of an herbicide tank mixture. For the CT system, metolachlor runoff was reduced 2x and pendimethalin 1.2x when compared with the non-irrigated treatment. The difference in irrigated and non-irrigated metolachlor means was significant (P = 0.05). Irrigation reduced metolachlor runoff by 1.3x in the ST system, but there was a 1.4x increase for pendimethalin. Overall results indicated that irrigation incorporation reduces herbicide runoff with the greatest impact when CT is practiced and products like metolachlor, which have relatively low K(oc) and high water solubility, are used. The lower ST system response was likely due to a combination of spray interception and retention by the ST system cover crop mulch and higher ST soil organic carbon content and less total runoff. During the study, the measured K(oc) of both herbicides on runoff sediment was found to vary with tillage and irrigation after herbicide application. Generally, K(oc) was higher for ST sediment and when irrigation incorporation was used with the CT system. These results have significant implications for simulation model parametization.

GLEAMS*, OPUS, AND PRZM-2 MODEL PREDICTED VERSUS MEASURED RUNOFF FROM A COASTAL PLAIN LOAMY SAND

Comprehensive models for agrichemical transport necessarily include runoff predictions to partition rainfall between infiltration and runoff, as this ability is fundamental to predictions of chemical runoff and leaching. We compared GLEAMS, Opus, and PRZM-2 model runoff predictions with runoff measured in a precisely controlled field site used for chemical runoff studies. In 1992 and 1993, two 14.5 m iA42.9 m corn (Zea mays, L.) field plots with 3% slope on Tifton loamy sand (fine-loamy, siliceous, thermic Plinthic Kandiudult) received six severe, artificial rainfall events over the growing season with each event consisting of a 25 mm h¨C1 rainfall for 2 h. Runoff was monitored continuously using a collector and flume. Model performance criteria included sensitivity analysis, graphical comparison and statistical analysis including mean, ratio of means, root mean square error (RMSE), and a paired difference t-test. Observed runoff averaged 20% of added rainfall. Lowest values occurred with freshly plowed soil or full canopy cover, while 24 to 34% runoff occurred when nearly bare soils had crusted over. Using an initial moisture condition-II curve number (CN) of 85, GLEAMS and Opus predicted runoff within 10%, overall, and produced a pattern of high and low runoff that closely followed observed. PRZM-2 overpredicted runoff by 90%, overall, and predicted its highest runoff when observed runoff was lowest. Paired difference t-tests indicated a significant difference between measured and predicted runoff for PRZM-2 (p<0.001 at ¦A= 0.05), but none for GLEAMS (p = 0.761) or Opus (p = 0.194). Mean, ratio of means, and RMSE showed that GLEAMS and Opus performed better than PRZM-2. All three models were very sensitive to CN values which were empirical and subjective, but less sensitive to measurable soil physical properties. With careful parameterization, GLEAMS and Opus could be used to simulate runoff from similar row-crop and soil conditions.

Tillage, cover-crop residue management, and irrigation incorporation impact on fomesafen runoff.

Intensive glyphosate use has contributed to the evolution and occurrence of glyphosate-resistant weeds that threaten production of many crops. Sustained use of this highly valued herbicide requires rotation and/or substitution of herbicides with different modes of action. Cotton growers have shown considerable interest in the protoporphyrinogen oxidase inhibitor, fomesafen. Following registration for cotton in 2008, use has increased rapidly. Environmental fate data in major use areas are needed to appropriately evaluate risks. Field-based rainfall simulation was used to evaluate fomesafen runoff potential with and without irrigation incorporation in a conventional tillage system (CT) and when conservation tillage (CsT) was practiced with and without cover crop residue rolling. Without irrigation incorporation, relatively high runoff, about 5% of applied, was measured from the CT system, indicating that this compound may present a runoff risk. Runoff was reduced by >50% when the herbicide was irrigation incorporated after application or when used with a CsT system. Data indicate that these practices should be implemented whenever possible to reduce fomesafen runoff risk. Results also raised concerns about leaching and potential groundwater contamination and crop injury due to rapid washoff from cover crop residues in CsT systems. Further work is needed to address these concerns.

Influence of tractor wheel tracks and crusts/seals on runoff: Observations and simulations with the RZWQM

In simulations on the fate of agricultural chemicals applied to crops, accurate partitioning of rainfall between infiltration and runoff is fundamental to chemical runoff predictions. We evaluated the Root Zone Water Quality Model (RZWQM version 3.1) against measured runoff from two field plots (15×45 m with 3% slope) on a Tifton loamy sand (fine-loamy, siliceous, thermic Plinthic Kandiudult). Six simulated rainfall events, each 25 mm h−1 for 2 h, were applied to maize (Zea mays, L.) each year. In the uncalibrated mode, RZWQM under-predicted runoff by 40% on average, with the closest fit for events that occurred after full canopy. Saturated hydraulic conductivity (Ks) accounted for the majority of the uncertainty in predicted runoff. When Ks of the surface crust was back calibrated from the measured runoff, RZWQM predicted runoff closely for the remaining plots and events. Alternatively, using different Ks values for wheel track and crop beds, running the model for each and, then, proportionally assigning runoff also led to predictions that agreed with measured runoff. When spatial and temporal changes in Ks were calibrated to specific conditions at the site, RZWQM effectively predicted runoff.

Agrichemical Transport to Groundwater through Coastal Plain Soils

A 1–ha field with Pine Flat loamy sand (coarse–loamy, siliceous, thermic Typic Paleudult) and Troup loamy sand (loamy, siliceous, thermic Grossarenic Kandiudult) surface soils, located near Plains, Georgia, was studied for four years (1993 to 1996) to evaluate potential agrichemical transport to groundwater. The field was managed to produce summer corn and winter wheat. Commercial fertilizer, the herbicide atrazine, and the insecticide carbofuran were applied to the field in 1993, 1994, and 1995. Average annual application rates were 266 kg nitrogen ha–1, 2.5 kg atrazine ha–1, and 2.4 kg carbofuran ha–1. Monthly soil–water and groundwater samples were collected. The samples were analyzed for nitrate nitrogen (NO3 ––N), chloride, atrazine, carbofuran, and deethylatrazine (DEA). Soil–water and groundwater samples indicated elevated NO3 ––N concentrations (>5 ppm) in the vadose zone at 4.3 m and in the aquifer at 10 m (>4 ppm). Of the studied pesticides, carbofuran and DEA were observed at the greatest concentrations in groundwater. Both NO3 ––N and pesticides were transported during groundwater recharge following periods of excess precipitation. Peak pesticide concentrations in groundwater were observed in late 1994, driven by a large precipitation event in July of 1994 when 565 mm of rain fell over a 4–day period. Atrazine and carbofuran concentrations in groundwater did not exceed the EPA maximum contaminant levels of 3 ppb and 40 ppb, respectively. Spatially averaged concentrations observed in monthly groundwater collected directly below the field were well below these standards. Concentrations of NO3 ––N, atrazine, DEA, and carbofuran observed in groundwater from the on–field wells were significantly different from up–gradient and down–gradient concentrations (p = 0.05). These data indicate a significant impact to the local groundwater. Nitrate N was transported down–gradient from the field at the largest concentrations. Peak concentrations of atrazine and DEA were simultaneously observed in the groundwater, indicating similar transport rates for both compounds and rapid transformation from atrazine into DEA in the root–zone.

Fenamiphos losses under simulated rainfall: Plot size effects

The purpose of this study was to compare two commonly used runoff experimental methods, which have different scales, on measurements of runoff and associated fenamiphos and metabolite losses over a 2-year period. Methods used were 15 m wide by 43 m long (645 m2) mesoplots and 1.8 m wide by 3 m long (5.4 m2) microplots, under simulated rainfall (25 mm h-1 for 2 h) at 1, 14, and 28 d after fenamiphos application. Mesoplots and microplots were established parallel to a 3% slope on a Tifton loamy sand (Plinthic Kandiudult). All plots were planted to corn (Zea mays L.). Target application rate for fenamiphos was 6.7 kg ha-1. Runoff totals and maximum rates for meso- and microplots were similar, with approximately 25% of the rainfall running off mesoplots and approximately 28% running off microplots. Runoff totals and maximum rates from meso- and microplots were each positively correlated (R2 = 0.89). In both years, fenamiphos lost in runoff decreased with each rainfall event (1, 14, and 28 d after application). The majority of fenamiphos lost in runoff was in the fenamiphos sulfoxide form. Fenamiphos sulfoxide lost over both years from mesoplots ranged from 51% to 93% of the total fenamiphos lost, and loss from microplots ranged from 47% to 100% of the total fenamiphos lost. Runoff from meso- and microplots 1 d after fenamiphos application, a “reasonable worst-case” event, had the greatest fenamiphos losses among events. Total losses of fenamiphos for this event averaged 1.2% (CV = 26%) of applied amount for mesoplots and 1.3% (CV = 47%) of applied amount for microplots. Maximum (seasonal) fenamiphos losses for meso- and microplots were 1.4% of applied for mesoplots and 2.6% of applied for microplots. A positive correlation was obtained between microplots and mesoplots for total losses of fenamiphos + metabolites (R2 = 0.88), fenamiphos parent (R2 = 0.89), and fenamiphos sulfoxide (R2 = 0.81). Relatively poor agreement was found for relatively small losses of fenamiphos sulfone between plot types (R2 = 0.34). Microplots and mesoplots yielded statistically similar results in terms of runoff and fenamiphos losses; thus, microplot results can be extrapolated up to larger mesoplot areas under these conditions. This has implications for field-scale management and watershed assessment in the Coastal Plain region of the southeast U.S. in that microplot and rainfall simulation results could be useful as statistically valid input datasets to estimate runoff and associated fenamiphos losses from larger areas.

ATRAZINE AND CARBOFURAN TRANSPORT THROUGH THE VADOSE ZONE IN THE CLAIBORNE AQUIFER RECHARGE AREA

A 1-ha field plot with a sandy surface soil, located near Plains, Georgia, was studied for three years (from 1993 to 1995) to evaluate pesticide transport in the vadose zone. Vadose zone soil samples were collected 23 times: prior to the initial 1993 pesticide application, each year at approximately 1, 3, 7, 14, 28, and 44 days after pesticide application, each fall after harvest, and in the spring of 1995 prior to planting. The samples were analyzed for atrazine, carbofuran, deethylatrazine (DEA), and deisopropylatrazine (DIA). Atrazine and carbofuran in the active root zone (< 100 cm) degraded rapidly. Overall, the higher concentration levels of atrazine, DEA, DIA, and carbofuran were limited to the top 25 cm of the profile and to the period from 1 to 30 days after application. On the average, by 30 days after application 83% of the atrazine and 96% of the carbofuran had degraded. By 44 days after application, virtually all of the pesticides in the top 250 cm of the soil had degraded. Atrazine was found to be more persistent than was carbofuran with a half life approximately twice that for carbofuran. A two-stage model with a variable dissipation rate for the period up to 44 days after pesticide application and a second dissipation rate for periods greater than that was found to fit the data better than a single stage model. For the first 44 days after application, the first-order decay rate with a half life of 12 days was found to fit the field data for atrazine within the soil profile. A first-order decay rate with a half life of approximately 6 days fit the observed carbofuran data best. The dissipation rate decreased rapidly after the first 44 days. When a two-stage dissipation process was assumed, the dissipation rate coefficient decreased from 0.059 to 0.006 (days -1 ) for atrazine, while for carbofuran it decreased from 0.110 to 0.018 (days -1 ). Observed levels of the atrazine metabolites DIA and DEA were highest in the top 1 cm of the soil. There appeared to be some movement or creation of the metabolites at lower depths in the profile later in the growing season, but not at large concentrations. Keywords. Soils, Aquifers, Pesticide transport, Water quality, Atrazine, Carbofuran.

Seasonal Hydrologic Impacts of Conservation Tillage for a Coastal Plain Soil

Strip till, a practice of planting into a narrowly tilled strip, is a growing practice among many Coastal Plain cotton growers. Strip tillage increases crop residue at the surface, leading to reduced evaporation and reduced raindrop impact. In many cases increased infiltration, reduced surface runoff, and reduced transport of sediment and agrichemicals are also observed. This research examines nine years of rainfall-runoff data from a paired conventional till / strip till research site. Annual water gains in the strip till system obtained through enhanced infiltration and reduced surface runoff were offset by increased subsurface losses. Strip till had the greatest benefit in terms of increased water gains during years with the least annual precipitation. These results indicate that strip till systems can potentially increase plant available water through enhanced infiltration. This increase is most prevalent during the crop growing season from June through August. Water losses through subsurface flow tend to cancel out any gains obtained through increased infiltration occurring throughout the remainder of the year. On an annual basis, the total water lost as a % of rainfall from the conventional till plots averaged 34% while it averaged 33% for the strip till plots. Of this, the conventional till plots lost 23% through surface runoff and 11% through subsurface losses. For the strip till plots the annual losses were 14% through surface runoff and 19% through subsurface losses.