Thomas A. Costello

Performance of Vegetative Filter Strips with Varying Pollutant Source and Filter Strip Lengths

Vegetative filter strips (VFS) can reduce runoff losses of pollutants such as nitrogen (N) and phosphorus (P) from land areas treated with fertilizers. While VFS effectiveness is considered to depend on lengths of pollutant source and VFS areas, there is little experimental evidence of this dependence, particularly when the pollutant source is manure-treated pasture. This study assessed the effects of pollutant source area (fescue pasture treated with poultry litter) length and VFS (fescue pasture) length on VFS removal of nitrate N (NO3-N), ammonia N (NH3-N), total Kjeldahl N (TKN), ortho-P (PO4-P), total P (TP), total organic carbon (TOC), total suspended solids (TSS), and fecal coliform (FC) from incoming runoff. This research examined poultry litter-treated lengths of 6.1, 12.2, and 18.3 m, with corresponding VFS lengths of up to 18.3 m, 12.2 m, and 6.1 m, respectively. Runoff was produced from simulated rainfall applied to both the litter-treated and VFS areas at 50 mm/h for 1 h of runoff. Pollutant concentrations in runoff were unaffected by litter-treated length but demonstrated a first-order exponential decline with increasing VFS length except for TSS and FC. Runoff mass transport of NH3-N,TKN, PO4-P, TP and TOC increased with increasing litter-treated length (due to increased runoff) and decreased (approximately first-order exponential decline) with increasing VFS length when affected by VFS length. Effectiveness of the VFS in terms of NH3-N, TKN, PO4-P, TP and TOC removal from runoff ranged from 12-75, 22-67, 22-82, 21-66, and 8-30% respectively. The data from this study can help in developing and testing models that simulate VFS performance and thus aid in the design of VFS installed downslope of pasture areas treated with animal manure.

IN SITU MEASUREMENT OF AMMONIA VOLATILIZATION FROM BROILER LITTER USING AN ENCLOSED AIR CHAMBER

An experimental apparatus and analytical technique were developed for measuring and characterizing in situ ammonia (NH3) volatilization from broiler litter. The apparatus included an enclosed air chamber, an infrared gas analyzer, a dew point hygrometer, and a datalogger. Airspeed in the chamber was within 12% of that measured in a typical broiler house. Ammonia flux was monitored during growout of six flocks of broilers where initial bedding material was either rice hulls or an equal mixture of rice hulls and pine shavings. Ammonia flux from re-used litter was six times that from new bedding material at the start of a growout. Mean and maximum NH3 flux during growouts were 149 and 314 mg NH3-N/m2-h on new bedding and 208 and 271 mg NH3-N/m2-h on re-used litter, respectively. Random experimental error for the flux measurement technique was estimated to be 32 mg NH3-N/m2-h. Systematic variation was found in transverse measurements across the width of a broiler house with flux greatest adjacent to the water lines. These results may be useful in designing economical strategies to control NH3.

RUNOFF FROM COTTON FIELDS FERTILIZED WITH POULTRY LITTER

Broiler litter is a valuable soil amendment for crop production, with litter typically applied to land areas used for grazing and/or hay production. Historically, farmers have based litter application rates on the nitrogen (N) needs of the receiving crop; however, this results in over–application of phosphorus (P). To alleviate the tendency to accumulate P in nearby watersheds, export of litter to other regions has been proposed. A field study was conducted at the University of Arkansas Northeast Research and Extension Center at Keiser, Arkansas, to quantify the impact of poultry litter on runoff water quality from cotton (Gossypium hirsutum L.) cropland in the Mississippi River Delta. Six 0.6–ha fields were fertilized with either commercial fertilizer (annually 140 kg N ha –1 yr –1 in a split application) or poultry litter (7,200 to 9,200 kg litter ha –1 yr –1 ). Runoff from each field was diverted through an H–flume, where water samples were collected. Stage was recorded and used to compute runoff flow. Concentration and mass loss of eight water–quality analytes were measured from 21 individual runoff events over a three–year monitoring period. Seedcotton yield was lower from the litter–fertilized plots, suggesting a need for refinements in the production system. Three–year total mass losses of total suspended solids (TSS) and nitrate nitrogen (NO3–N), as well as total runoff volume, were reduced from fields receiving poultry litter compared to fields receiving commercial fertilizer. Differences associated with the two fertilizer treatments did not clearly exhibit a systematic change over three years, as would be expected if soil structure were changing as a result of the organic amendments. Although total TSS mass loss was reduced, concentrations of nutrients associated with the solid fraction of chicken litter increased periodically in runoff events that immediately followed litter application. Moreover, total loss of orthophosphate P from litter–treated fields was four times greater than from fields fertilized conventionally. Additional research is needed to ensure that poultry litter exported to the Mississippi River Delta region can be applied as part of an integrated crop fertility program that results in the desired agronomic performance while protecting and enhancing water quality.

Pyrolysis kinetics of algal consortia grown using swine manure wastewater.

In this study, pyrolysis kinetics of periphytic microalgae consortia grown using swine manure slurry in two seasonal climatic patterns in northwest Arkansas were investigated. Four heating rates (5, 10, 20 and 40 °C min(-1)) were used to determine the pyrolysis kinetics. Differences in proximate, ultimate, and heating value analyses reflected variability in growing substrate conditions, i.e., flocculant use, manure slurry dilution, and differences in diurnal solar radiation and air temperature regimes. Peak decomposition temperature in algal harvests varied with changing the heating rate. Analyzing pyrolysis kinetics using differential and integral isoconversional methods (Friedman, Flynn-Wall-Ozawa, and Kissinger-Akahira-Sunose) showed strong dependency of apparent activation energy on the degree of conversion suggesting parallel reaction scheme. Consequently, the weight loss data in each thermogravimetric test was modeled using independent parallel reactions (IPR). The quality of fit (QOF) for the model ranged between 2.09% and 3.31% indicating a good agreement with the experimental data.

Estimating pan evaporation using limited meteorological observations

Abstract Pan evaporation measurements have been used worldwide as a means of estimating evapotranspiration and free water surface evaporation. The availability of pan evaporation measurements is critical to many applications including irrigation scheduling, hydrologic modeling, and irrigation system design. The ability to predict daily pan evaporation based on limited meteorological observations is highly desirable. In many situations it is advantageous to calculate, rather than measure, pan evaporation. This is often the case in developing nations or remote locations where costs are prohibitive. Many relationships that predict pan evaporation or potential evapotranspiration as a function of limited meteorological observations have been developed. The relationships are often empirical, thus they suffer from lack of transferability. The relationship between potential evapotranspiration and pan evaporation facilitates local calibration of established equations using measured pan evaporation data. This calibration may then be extrapolated to locations that have similar climatic characteristics. A calibration procedure is presented and demonstrated by calibrating three equations using data from several sites located in the southeastern United States.

A modified Weibull model for growth and survival of Listeria innocua and Salmonella Typhimurium in chicken breasts during refrigerated and frozen storage

The potential of food-borne pathogens to survive and grow during refrigerated and frozen storage has raised serious concerns over the safety of stored poultry products. In this study, the effect of refrigeration and freezing temperatures (-20, -12, 0, 4, and 8°C) on growth and survival of Listeria innocua and Salmonella enterica serovar Typhimurium in raw chicken breasts for storage times of 3, 7, 10, 14, and 21 d were investigated. A modified Weibull model was also developed to analyze the microbial behavior of both microorganisms in raw chicken breasts under different refrigerated storage conditions over time. The results showed that the bacterial loads of L. innocua at 4 and 8°C and Salmonella Typhimurium at 8°C were significantly different (P < 0.05) from those at other refrigerated and frozen storage temperatures over storage times. The loads of both bacteria at frozen storage temperatures did not change significantly over time. At a storage time of 7 d, the increase in bacterial loads of L. innocua at 4 and 8°C was 2.1 log cfu/g and 3.7 log cfu/g, respectively, and that of Salmonella Typhimurium at 8°C was 1.2 log cfu/g. The root mean square errors, median relative error, mean absolute relative error, and the plot of predicted versus observed bacterial loads showed a good performance of the model. The results from this study provided useful information regarding the behavior of Listeria and Salmonella in raw chicken breast meat during refrigerated and frozen storage, which would be helpful in giving insight over the safety of poultry products storage.

NUTRIENT RETENTION, NUTRIENT LIMITATION, AND SEDIMENT-NUTRIENT INTERACTIONS IN A PASTURE-DOMINATED STREAM

Even though agricultural land use is a principal source of nonpoint-source pollution, the in-stream movement and transport of nutrients and their interaction with benthic sediments are not well understood. The objectives of this study were to: (1) assess if nutrients were limiting algal growth, (2) assess equilibrium between sediments and water column phosphorus (P), (3) evaluate spatial and temporal variability in P buffering capacity and easily exchangeable P, and (4) evaluate variations in stream nutrient retention efficiency in Moores Creek, a pasture-dominated watershed in northwest Arkansas. This study used nutrient limitation bioassays, sediment nutrient extractions, and short-term nutrient injections to accomplish the study objectives. Results suggested that light, not nutrients, was limiting algal growth in Moores Creek. Sediment equilibrium P concentration measurements suggested that sediments may be releasing dissolved inorganic P during winter and spring; however, sediments were a potential temporary sink of dissolved inorganic P during summer and fall. Exchangeable P concentrations varied between 0.4 and 1.0 mg kg-1 of dry sediments, and P sorption index varied between 3.2 and 5. Concentrations of ammonia nitrogen (NH4-N) and ortho P (PO4-P) generally decreased between sampling stations during nutrient injections at all three sites; however, nitrate N (NO3-N) concentrations increased downstream. A significant retention of NO3-N was not observed in any of the experiments, indicating that agricultural headwater streams may be a source of downstream NO3-N transport. The results indicated that a riparian cover should be maintained in nutrient-rich headwater streams in order to minimize algal production in the streams. Even nutrient-rich streams may continue to assimilate, to some extent, increased loads of P, altering the timing and magnitude of downstream transport of P.

VALIDATING A VEGETATIVE FILTER STRIP PERFORMANCE MODEL

Vegetative filter strips (VFS) reduce losses of nutrients, solids, and other materials from land area treated with fertilizers and manures. A number of models are available that simulate nutrient and sediment transport in VFS. While VFS effectiveness is considered to depend on lengths of pollutant source area and VFS areas, few published studies have tried to validate these models using variable pollutant source area and VFS area. The objective of this study was to validate an event-based nutrient transport model (Chaubey et al., 1995) that simulates soluble nutrient transport in VFS. This model links three sub-models: modified Green-Ampt infiltration, non-linear kinematic wave overland flow routing, and a nutrient transport component. The nutrient transport component considers infiltration as the only mechanism of pollutant removal from runoff. Data from a field plot experiment were used to validate the model. The model was executed using an uncalibrated runoff component, a calibrated runoff component, and measured runoff. The concentrations of parameters entering the VFS from three different poultry litter application lengths (6.1, 12.2, and 18.3 m) were not significantly different. However, predicted concentrations at subsequent lengths were different for all the three poultry litter application lengths. This finding was consistent with the observed data. Model execution with the uncalibrated runoff component, calibrated runoff component, and measured runoff underpredicted concentrations and mass transport at various locations along the length of the VFS. Underprediction of concentration was judged to be the reason for underprediction of mass transport. The agreement between the observed and predicted concentrations and mass transport, however, improved when runoff predictions from the calibrated runoff component and measured runoff were used. This suggests that accurate prediction of infiltration and runoff is critical for accurate prediction of concentration mass transport. Furthermore, since concentration was underpredicted even when measured runoff was used, this study suggests that the nutrient transport component might be improved, possibly by including nutrient removal mechanisms other than infiltration.

Effect of Rice Moisture Content at Harvest on Economic Return

Economic analyses were conducted to assess the impact of rice harvest moisture content (MC) on economic return to a producer. The primary factors considered included percent milled rice, percent head rice, field yield, and drying costs. The results showed that rice harvest MC had a dramatic impact on the final gross income to a producer. The optimal harvest MC was influenced by field yield loss rate and, to a lesser degree, drying charge schedule. The relative price of broken to head rice also affected the optimal harvest MC. Significant losses in gross income could be incurred if rice is harvested at MCs lower than 15% or higher than 22%.

Modeling Long-grain Rice Milling Quality and Yield During the Harvest Season

Three mathematical models were developed to predict percent milled rice (PMR), percent head rice (PHR), and field yield (FY) of long-grain varieties as a function of rice moisture content (m.c.) at harvest. The concept of potential yield was used in the model development. The actual yield was equal to the potential yield minus the yield reduction caused by specific weather conditions. The PMR and PHR models were validated using experimental data collected for three long-grain rice varieties from 1987 to l990. The potential FY model was validated using the data reported by Morse et al. (1967) and Bal and Ojha (1975). Predicted PMRs compared well with the experimental data with the average prediction errors of less than 1.9 percentage points. The average differences between the predicted and measured PHRs were less than 2.4 percentage points when rice was harvested between 13 and 24% m.c. Large prediction errors (more than 5 percentage points) were obtained in limited cases where rice was harvested at m.c.s greater than 24% or less than 13%. The predicted FYs compared reasonably well with the experimental data by Morse et al. (1967) and Bal and Ojha (1975) with the average errors of 3.5 and 6.8%, respectively.