HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Schwartz, L. Articles by Shuman, L. M. Search for Related Content PubMed PubMed Citation Articles by Schwartz, L. Articles by Shuman, L. M. Agricola Articles by Schwartz, L. Articles by Shuman, L. M. Related Collections Watershed and Landscape Processes Surface Water Quality Nutrients Solute Transport Models Runoff

TECHNICAL REPORTS

Surface Water Quality

Predicting Runoff and Associated Nitrogen Losses from Turfgrass using the Root Zone Water Quality Model (RZWQM)

Department of Crop and Soil Sciences, University of Georgia, 1109 Experiment Street, Griffin, GA 30223

^* Corresponding author (schwartz{at}griffin.uga.edu)

Received for publication December 10, 2003.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Nitrogen fertilizers are used to maintain optimum turfgrass quality, but off-site movement of this primary nutrient can affect water quality. We conducted a 4-yr study (1998–2001) designed to measure nitrate N runoff from turfgrass, gathering data to be used in the model. The process-based Root Zone Water Quality Model (RZWQM) was used to predict nitrate associated with runoff from turfgrass. The measurements were made on 12 sloped (5%), 25-m² plots of ‘Tifway’ bermudagrass [Cynodon dactylon (L.) Pers.], managed as golf course fairways and exposed to natural and simulated rainfall. Surface runoff volume and nitrate N loads were monitored after applying simulated rainfall at an average intensity of 27.4 mm h^–1. The irrigation occurred (25 or 50 mm) 4 through 168 h after treatment with various rates of N fertilizers for 1998–2001. RZWQM adequately simulated water runoff volumes (<19%; nonsignificant, paired t test) in the first three (normal or wet) years, but overpredicted (70%) in the fourth, dry year. RZWQM overpredicted nitrate N loads by a factor of 1.3 for the first three years (nonsignificant), and by a factor of almost 6 for the fourth year (highly significant). These overpredictions occurred when the runoff volumes and N loads were very small. The research has shown that refinements to RZWQM are needed for turfgrass management applications.

Abbreviations: RZWQM, Root Zone Water Quality Model

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
CONTAMINATION OF ground and surface water systems by agricultural chemicals is a global concern. Researchers and regulatory agencies are investigating the sources of contamination and trying to ameliorate them through implementing best management practices (BMPs). One possible source of pollution is turfgrass, one of the most intensively managed land uses in the United States (King et al., 2001; Banks et al., 2004). The USEPA estimated that there are between 6.5 and 9.7 million hectares of maintained turfgrass in the United States (Liskey, 1997). Turfgrass management requires high inputs of fertilizers, herbicides, and other pesticides (Shuman, 2002; Easton and Petrovic, 2004). Therefore, it is essential to understand the environmental impact of these chemicals on water quality.

Large-scale field monitoring studies are time consuming, expensive, and difficult to conduct, being mostly dependent on natural rainfall. Often, long periods of time elapse between severe runoff events (Ma et al., 1999). As a consequence, modeling represents an important tool in studying the influences of various management practices on water quality. Modeling the effects of nutrient losses on environmental resources such as ground and surface water can help in finding alternative best management practices to reduce these losses. Even though modeling has been widely used for agricultural systems, few modeling studies have been published for urban settings such as golf courses. For instance, LEACHM (Hutson and Wagenet, 1992) has been used to simulate nitrate N and ammonium N content in Cecil (fine, kaolinitic, thermic Typic Kanhapludult) soil horizons as well as leached nitrate concentrations in subsurface drainage (Johnson et al., 1999). It was shown that the model generally underestimated nitrate N and ammonium N in soil during the winter and overestimated soil ammonium N during the summer. The model also overestimated cumulative drainage and leached nitrates during both seasons.

Other models have been designed to simulate surface water runoff and pesticide transport from turfgrass, such as PRZM-2 (Mullins et al., 1993; Cohen et al., 1993), SWRRBWQ (Arnold et al., 1991; Cohen et al., 1993), GLEAMS (Leonard et al., 1987), and OPUS (Smith, 1992; Ma et al., 1999). Studies conducted on four golf courses in the northeastern United States with five pesticides have shown that PRZM-2 reasonably simulated pesticide concentrations in ground water. The SWRRBWQ model was used to guide pesticide selection on golf courses in Hawaii by comparing predicted with measured concentrations in runoff water. The GLEAMS model, when used for predicting the leaching of 2,4-D (2,4-dichlorophenoxyacetic acid) on golf course greens, greatly overestimated leaching of the pesticide. The OPUS model was used in analyzing the results from a 3-yr field study on bermudagrass by measuring surface runoff and various pesticides losses in the runoff. The model adequately simulated the runoff but not the pesticide losses.

The mixed success of the previously mentioned models prompted us to study the applicability of another model, RZWQM, for predicting nutrient runoff from golf courses. RZWQM is a process-based, numerical, one-dimensional model that has components for runoff and chemical transfer to runoff. It simulates the movement of water, nutrients, and pesticides through and over the soil on a field scale as affected by soil properties, weather, agrochemical applications, and crop and agricultural management practices (Rojas et al., 1988; Ahuja et al., 1996, 2000).

Shuman (2002)(2004) measured nitrate N from simulated golf course fairways of Tifway bermudagrass. In these studies, the ultimate goal was to determine management practices that minimize N transport to surface waters. Nitrogen can become a serious pollutant, depending on its form and quantity, if it is lost to watercourses or deep ground water. Mallin and Wheeler (2000) stated that nitrate N levels as low as 0.1 mg L^–1 can cause eutrophication. Typical N fertilizers, being water soluble, are very much influenced by the runoff water quantity. Thus, it is expected that higher losses of the N fertilizers will occur when higher amounts of runoff will be present. Moreover, a study conducted by Fleming and Cox (2001) on dairy pastures (similar, therefore, to turfgrass) in South Australia indicated that runoff represents up to 10% of annual rainfall, and more than 90% of the runoff was overland flow as opposed to horizon interflow.

Fertilizer application rate will be another factor that influences nitrate N loading in runoff. In that regard, significant differences exist between agronomic crop and turfgrass plots. For the U.S. Southeast, the typical recommended application rate for corn (Zea mays L.) production is 134 kg N ha^–1, whereas for turfgrass, it is 336 kg N ha^–1 (Plank, 1989, p. 153–156), or 2.5 times more on an annual basis.

In this study, we tested RZWQM to predict nitrate N losses in runoff from small field plots over 4 yr. The model was calibrated for a two-year period, 1998–1999, an essential first step in the basic protocol for a hydrological model (Mulla and Addiscott, 1999). Calibration serves as a tool to estimate those parameters that cannot be easily measured or determined (Hanson et al., 1999). Traditionally, when a model has to be calibrated and validated, one set of data is used for calibration and an independent set is used for testing. Therefore, after calibration, RZWQM was validated for runoff and nutrient transport from turfgrass for the two-year period, 2000–2001.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Summary of Field Study
Twelve individual plots (7.0 x 3.6 m) of Tifway bermudagrass separated by landscape timbers were built in a grid on a Piedmont soil of the southeastern United States (Shuman, 2002, 2004). All plots had complete ground cover, 5% slope, and were exposed to natural rainfall and managed as golf course fairways. The topsoil was a Cecil sandy loam that has a mixed surface horizon (Table 1). The soils are deeply weathered, kaolinitic, acidic, and variably charged. A trough was installed in a ditch at the lower end of each plot to collect the runoff. A tipping bucket flow meter measured runoff volume and overhead sprinklers produced simulated rainfall at intensity of 27.4 mm h^–1.

Meteorological data were obtained from the weather station at the Griffin campus of the University of Georgia. The cumulative precipitation for 1998–2001 including natural and simulated rainfall was 6710 mm.

Samples collected from each rainfall were stored at 4°C before analysis. Nitrate N was determined by filtering samples through 0.45-µm filters, followed by colorimetric analysis using a Lachat (Milwaukee, WI) flow analyzer.

RZWQM Description and Application
RZWQM is a one-dimensional model that provides a comprehensive analysis of root zone processes that affect water quality as affected by a wide range of agricultural management practices. It is an integrated physical, biological, and chemical process-based model that simulates plant growth and movement of water, available nitrogen, and plant production (Hanson et al., 1999). RZWQM simulates hydrological processes in the soil at sub-daily time steps (from 5–60 min), overcoming many of the difficulties encountered in daily time step models. Unlike other runoff models, RZWQM calculates runoff in "real time," as hydrologic events occur (Ahuja et al., 2000). During a rainfall event, water seeps into the profile following the infiltration relationship of Green and Ampt (Ahuja and Hebson, 1992). When rainfall intensity exceeds the computed infiltration rate, the excess is allowed to flow into macropores, if present. If macropores are not present, or if they are filled, the remaining excess leaves the field as runoff. The amount of water required to satisfy macropore flow in an infiltration time step is calculated from the maximum flow-rate capacity of the macropores, which is calculated using Poiseuille's law and assuming gravity flow. In this study we have used the default values used under the "Soil Physical Parameters" section of the model for the specific type of soil and for turfgrass.

The soil water parameters of Brooks–Corey play an important role in calculations of runoff (Ahuja et al., 2000). The water retention data for Cecil sandy clay loam were taken from a study done at the Griffin campus of the University of Georgia (Ma et al., 1999). It has been shown that the sum of square errors for the curve fitting was on the order of 10^–4 for soil water content, indicating that the modified Brooks–Corey parameters adequately described the measured water retention data. Other measured soil and soil hydraulic properties of the Cecil sandy clay loam, such as soil water content at 33 kPa suction head and soil water content at 1500 kPa suction head, were available. Initial soil water content was set to increase with depth and varied from 0.3 to 0.43 cm³ cm^–3 (Ma et al., 1999).

RZWQM, similar to other models, considers movement of N from applied fertilizer by solute transport, uptake of N by plants, mineralization and immobilization (between organic and inorganic forms), nitrification–denitrification, and volatilization of ammonium. To estimate chemical losses in runoff, RZWQM uses a simpler version of the nonuniform mixing model of Ahuja and Heathman (Ahuja et al; 2000). In that model, starting at the time of runoff initiation, rainwater and surface soil undergo nonuniform but equilibrium mixing over small time steps in 1-mm depth increments within the top 2 cm of soil. RZWQM applies an average degree of mixing to the top 1-cm layer, after which the average degree of mixing in each of the mm-thick layers decreases exponentially with the depth below the surface (Ahuja et al., 2000). The results are numerically integrated over this depth to obtain chemical transfer to runoff. RZWQM has an interactive procedure that allows the user to cycle through the scenario many times. Each time, initial conditions are reset to the conditions at the end of the previous run so that soil microbial, organic matter, and nutrient pools equilibrate and approach values appropriate for the agronomic conditions of the scenario. We cycled the model for 10 yr with local climatic data, site management practices, and parameterized soil properties. After these equilibrating runs, we compared the nitrate N loads in the runoff with those from the field.

Soil properties included measured particle size distribution, depth, soil horizon, and hydraulic conductivity (Table 1). Soil profile properties were available to a depth of 126 cm, divided into four horizons. Soil particle density was assumed to be 2.65 g cm^–3 and the pH varied from 5.8 to 4.8. Soil moisture content in the upper 7.5 cm, measured before adding simulated rainfall and before fertilizer application, was 16 to 27% (Shuman, 2002).

Climatic data were obtained from a nearby weather station. The meteorological data used daily values of minimum and maximum temperature, wind run, solar radiation, pan evaporation, and relative humidity (Georgia Environmental Monitoring Network, 2003). Local 15-min rainfall data were converted to breakpoint form (Ahuja et al., 2000).

Mean rainwater concentrations of NH₄⁺ (0.12 mg L^–1) and NO₃^– (0.50 mg L^–1) were taken from National Atmospheric Deposition Program data for 1998–2001 (National Atmospheric Deposition Program, 2004). Irrigation water N concentrations (0.1 mg L^–1 NH₄⁺ and 0.4 mg L^–1 NO₃^–) are typical of measured values for the area. Fertilizer data was taken from management records for the experiment (Table 2). Nutrient system C to N ratio and published soil chemical properties typical for similar systems were provided from personal communications with USDA-ARS scientists (R. Lowrance and R.K. Hubbard) in Tifton, GA.

The model was calibrated using the 1998–1999 data and then used to predict the 2000–2001 data for comparison. The calibration was done by adjusting two parameters found to be sensitive from preliminary model runs: the saturated conductivity (K_sat) of the bed's crust or seal and the field saturation fraction. The surface crust or seal influences runoff predictions as does the field saturation factor. In this study, the adjusted crust or seal was 0.2 cm h^–1, and the field saturation factor was set at 0.94 (Ma et al., 1998). Using an iterative process, we then slightly adjusted values for nutrient system C to N ratios, initial soil nitrate N, and ammonia N content until the nitrate N loads in the runoff were simulated as closely as possible to measured values. The subsequent 2-yr period served for comparing the predicted with the measured runoff volumes.

The methods for evaluating the goodness of fit of model predictions were the percentage difference between measured and predicted values [(measured – predicted)/measured x 100] and the paired difference t test. For each plot we measured the differences between measured and predicted values, then summed the differences and tested whether the obtained number was different from zero. The paired difference t test is appropriate for testing agreement between measurements and model predictions using linear regression analysis (SAS Institute, 2000), in which a p value was calculated and the significance of the differences was evaluated.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Hydrology
Turfgrass creates a unique environment in the sense that dense vegetation and thatch buildup can both enhance water infiltration when inputs are low-to-moderate, but may partially impede water movement into the soil when natural precipitation or sprinkler irrigation intensity is higher than infiltration capacity (Taylor and Blake, 1982). This factor, plus an existing soil crust (Ma et al., 1998), can cause lateral surface movement (runoff) of water and any soluble or suspended materials in it. Accurately predicting runoff volumes is important, though difficult, because this is expected to correlate strongly with loss of fertilizers, other agrochemicals, or sediment into surface waters. Runoff prediction is an important step in model validation and, implicitly, a tool to find and test management practices that mitigate losses.

RZWQM's response to simulated and natural rainfall, in terms of runoff generation, was examined for 4 yr (1998–2001). Observed and predicted amounts are shown in Fig. 1 through 4 . Figure 5 presents the cumulative natural and simulated rainfall as well as observed runoff for 1998–2001. As can be seen, the greatest amount of precipitation occurred in summer 1999 (during the first half of July); this is reflected in greater collected runoff then (Fig. 2 and Table 3), 16.9 cm. In contrast, the relatively low precipitation occurring in September and October 2001 is reflected by low amounts of collected runoff (Fig. 4 and 5 and Table 3). Thus, over the 4-yr period of this field experiment, all scenarios from low to normal to high precipitation were experienced.

View larger version (45K): [in this window] [in a new window]   Fig. 1. Measured versus calibrated runoff and nitrate N daily loads for 1998.

View larger version (31K): [in this window] [in a new window]   Fig. 4. Measured versus predicted runoff and nitrate N daily loads for 2001.

View larger version (34K): [in this window] [in a new window]   Fig. 2. Measured versus calibrated runoff and nitrate N daily loads for 1999.

View larger version (40K): [in this window] [in a new window]   Fig. 3. Measured versus predicted runoff and nitrate N daily loads for 2000.

View larger version (18K): [in this window] [in a new window]   Fig. 5. Cumulative rainfall and irrigation and runoff for 1998–2001.

Nitrate Loads in Surface Runoff
For grasses, fertilizer is most often applied in granular formulation (such as our 10–10–10 and ammonium nitrate–superphosphate treatments). In our experiments, fertilizer was applied to plots that had earlier received natural or artificial precipitation (so the soil was moist), followed after 4 h by additional irrigation. Experience shows that irrigation or rainfall, especially if intense, leads to some rapid transfer of nitrate from the granules into water that moves as runoff. It was hypothesized that N from more soluble fertilizers (e.g., 10–10–10) would experience greater runoff losses than from controlled-release fertilizers. That is why in the experiments, 10–10–10 fertilizer was applied at different rates to determine whether runoff losses were a linear function of rate, as well as to determine the effect on runoff of "watering-in" fertilizer nutrients (Shuman, 2002, 2004).

Nitrate N losses were monitored in surface runoff for 4 yr for natural and simulated rainfall events. The results related to the calibration and evaluation of measured versus predicted nitrate N loads in the specific runoff events, as well as their comparison, are presented in Fig. 1 through 4 and Table 3. Loading is a function of nitrate concentration and runoff volume: a higher amount of runoff loss induces a higher amount of nitrate N losses in runoff. Thus, the greatest amount of precipitation (natural and simulated rainfall) in the summer of 1999 (Fig. 2 and Table 3) caused a higher loss of nitrate N in the runoff (Fig. 2 and 6 , Table 3), 10.6 kg ha^–1 for the field measurements. At the opposite extreme, a lower quantity of runoff gave the lowest loss of nitrate N in runoff, 0.3 kg ha^–1, observed during fall 2001 (Fig. 4 and Table 3). This can be explained by the fact that lower precipitation (natural and simulated rainfall) causes drier soil. Unless applied very quickly, added water then infiltrates into the profile rather than running off. This effect was emphasized by the "watering-in" technique, used in 2001 for the first time in this 4-yr period. Not only did it provide a lower amount of simulated runoff, it also tended to move solubilized nitrogen into the upper layer of the soil, making it less susceptible to runoff losses.

View larger version (12K): [in this window] [in a new window]   Fig. 6. Predicted nitrate N lost to runoff during 1998–2001.

Overall, for the 2-yr calibration period the model showed good percentage difference values, being 12.5 and 21.7 for 1998 and 1999, respectively (Table 3). This conclusion is reinforced by the paired t test of measured versus calibrated values (Table 3), which indicated that those mean differences (0.425 and 0.446, for 1998 and 1999) were nonsignificant.

For the validation period, percentage difference indicates generally good agreement between measured and predicted values of the nitrate N loads in runoff for 2000, being 60% (Table 3). Nevertheless, the paired t value, 0.023, was significant at the 5% level, indicating a significant discrepancy between observed and predicted nitrate N loading amounts. This could be explained by the fact that we deal with low nitrate N loads, 0.5 and 0.8 kg ha^–1, respectively; error in measurements is expected to be greater, in such cases, as well as variability between plots.

In 2001, the model overpredicted the nitrate N loads by a factor of six (Table 3). We believe this relates to the lower runoff volumes in 2001. Variability in measured data will affect the accuracy of the predictions. Furthermore, as stated, "watering-in" was used in 2001, and this method, although successful in limiting nitrate N runoff loss, was apparently not well accounted for in the RZWQM assumptions. The overprediction is shown by the paired t test, in which the value for 2001, 0.004, was significant at the 1% level. Also, late in fall, the model tended to overpredict nitrate N loads in runoff (Fig. 4). This tendency might be explained by the fact that at lower temperatures, recycling of organically bound N in turfgrass had slowed, but the model has not been sufficiently refined to account for these complex biological processes.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
The results obtained show that the model generally described well both water and nitrate N loss from turfgrass in a scenario that simulated golf course fairways. An exception to this good predictability is related to the nitrate N loads simulated in 2001. Although low, the 2001 nitrate N runoff losses were substantially overestimated relative to observed values. This is attributable to several factors, one being that the very low volumes of runoff and nitrate N magnify the errors of the predictions. Very low runoff volumes and N loads are hard to measure accurately and there was a large variability among the replicate plots. Such low amounts are also hard to model accurately. The management protocol, "watering-in", used only in 2001, also is believed to have partially protected the surface-applied fertilizer from runoff loss. Furthermore, recycling of organically bound N during low temperatures is not sufficiently accounted for.

	ACKNOWLEDGMENTS

 
The authors gratefully acknowledge Ray Pitts, Kathy Evans, and Garland Layton for technical assistance. This research as supported by the United States Golf Association, the Georgia Turfgrass Foundation Trust, and by State and HATCH funds allocated to the Georgia Experiment Stations.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Ahuja, L.R., and C. Hebson. 1992. Water, chemical and heat transport in soil matrix and macropores. p. 1–26. In Root Zone Water Quality Model: Version 1.0. Technical documentation. GPSR Tech. Rep. 2. USDA-ARS-NPA, Fort Collins, CO.
• Ahuja, L.R., Q.L. Ma, K.W. Rojas, J.T.I. Boesten, and H.J. Farahani. 1996. A field test of Root Zone Water Quality Model—Pesticide and bromide behavior. Pestic. Sci. 48:101–108.
• Ahuja, L.R., K.W. Rojas, J.D. Hanson, M.J. Shaffer, and L. Ma. 2000. Root Zone Water Quality Model. Modeling management effects on water quality and crop protection. Water Resour. Publ., Highlands Ranch, CO.
• Arnold, J.G., J.R. Williams, R.H. Griggs, and N.B. Sammons. 1991. SWRRBWQ: A basin scale model for assessing management impacts on water quality. USDA-ARS in cooperation with Texas Agric. Exp. Stn. and Texas A&M Univ., Temple, TX.
• Banks, P.A., B. Branham, K. Harrison, T. Whitson, and I. Heap. 2004. Determination of the potential impact from the release of glyphosate- and glufosinate resistant Agrostis stolonifera L. in various crop and non-crop ecosystems [Online]. Available at www.wssa.net/wsinfo/pubs/bentgrass.pdf (verified 27 Aug. 2004). Weed Sci. Soc. Am., Lawrence, KS.
• Cohen, S.Z., T.E. Durborow, and N.L. Barnes. 1993. Groundwater and surface water risk assessments for proposed golf courses. p. 214–227. In K.D. Racke and A.R. Leslie (ed.) Pesticides in urban environments: Fate and significance. Am. Chem. Soc., Washington, DC.
• Easton, Z.M., and A.M. Petrovic. 2004. Fertilizer source effect on ground and surface water quality in drainage from turfgrass. J. Environ. Qual. 33:645–655.[Abstract/Free Full Text]
• Fleming, N.K., and J.W. Cox. 2001. Carbon and phosphorus losses from dairy pasture in South Australia. Aust. J. Soil Res. 39:969–978.
• Georgia Environmental Monitoring Network. 2003. Georgia weather—Automated Environmental Monitoring Network page [Online]. Available at www.griffin.peachnet.edu/bae/ (verified 27 Aug. 2004). Georgia Environ. Monitoring Network, Griffin, GA.
• Hanson, J.D., K.W. Rojas, and M.J. Shaffer. 1999. Calibrating the Root Zone Water Quality Model. Agron. J. 91:171–177.[Abstract/Free Full Text]
• Hutson, J.L., and R.J. Wagenet. 1992. LEACHM: Leaching estimation and chemistry model: A process-based model of water and solute movement, transformations, plant uptake and chemical reactions in the unsaturated zone. Res. Ser. 93-3. Cornell Univ., Ithaca, NY.
• Johnson, A.D., M.L. Cabrera, D.V. McCracken, and D.E. Radcliffe. 1999. LEACHN simulations of nitrogen dynamics and water drainage in an Ultisol. Agron. J. 91:597–606.[Abstract/Free Full Text]
• King, K.W., R.D. Harmel, H.A. Torbert, and J.C. Balogh. 2001. Impact of a turfgrass system on nutrient loadings to surface water. J. Am. Water Resour. Assoc. 37:629–640.
• Leonard, R.A., W.G. Knisel, and D.A. Still. 1987. GLEAMS: Groundwater loading effects of agricultural management systems. Trans. ASAE 30:1403–1418.
• Liskey, E. 1997. Research update. Grounds Maint. 32(3):54–55.
• Ma, Q.L., A.E. Smith, J.E. Hook, R.E. Smith, and D.C. Bridges. 1999. Water runoff and pesticide transport from a golf course fairway: Observations vs. Opus model simulations. J. Environ. Qual. 28:1463–1473.[Abstract/Free Full Text]
• Ma, Q.L., R.D. Wauchope, J.E. Hook, A.W. Johnson, C.C. Truman, C.C. Dowler, G.J. Gascho, J.G. Davis, H.R. Summer, and L.D. Chandler. 1998. Influence of tractor wheel tracks and crusts/seals on runoff: Observations and simulations with the RZWQM. Agric. Syst. 57:77–100.
• Mallin, M.A., and T.L. Wheeler. 2000. Nutrient and fecal coliform discharge from coastal North Carolina golf courses. J. Environ. Qual. 29:979–986.[Abstract/Free Full Text]
• Mulla, D.J., and T.M. Addiscott. 1999. Validation approaches for field-, basin-, and regional-scale water quality models. p. 63–78. In D.L. Corwin, K. Loague, and T.R. Ellsworth (ed.) Assessment of non-point source pollution in the vadose zone. Am. Geophys. Union, Washington, DC.
• Mullins, J.A., R.F. Carsel, J.E. Scarbrough, and A.M. Ivery. 1993. PRZM-2: A model for predicting pesticide fate in the crop root and unsaturated soil zones. Users manual for Release 2.0. USEPA Rep. 600/R-93/046. U.S. Gov. Print. Office, Washington, DC.
• National Atmospheric Deposition Program. 2004. National Atmospheric Deposition Program [Online]. Available at http://nadp.sws.uiuc.edu (verified 27 Aug. 2004). NADP, Champaign, IL.
• Plank, C.O. 1989. Soil test handbook for Georgia. Georgia Coop. Ext. Serv., Univ. of Georgia, Athens, GA.
• Rojas, K.W., C.S. Hebson, and D.G. DeCoursey. 1988. Modeling agricultural management subject to subsurface water quality constraints. p. 108–116. In Proc. Int. Symp. Modeling Agric., Forest and Hydrol. ASAE Winter Meeting, Chicago. 12–13 Dec. 1988. Am. Soc. Agric. Eng., St. Joseph, MI.
• SAS Institute. 2000. The SAS system for Windows. Release 7.0. SAS Inst., Cary, NC.
• Shuman, L.M. 2002. Phosphorus and nitrate nitrogen in runoff following fertilizer application to turfgrass. J. Environ. Qual. 31:1710–1715.[Abstract/Free Full Text]
• Shuman, L.M. 2004. Runoff of nitrate nitrogen and phosphorus from turfgrass after watering-in. Commun. Soil Sci. Plant Anal. 35:9–24.
• Smith, R.E. 1992. Opus: An integrated simulation model for transport of nonpoint-source pollutants at the field scale. Vol. 1. Documentation. USDA-ARS Rep. ARS-98. U.S. Gov. Print. Office, Washington, DC.
• Taylor, D.H., and G.R. Blake. 1982. The effect of turfgrass thatch on water infiltration rates. Soil Sci. Soc. Am. J. 46:616–619.[Abstract/Free Full Text]

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2005 34: 1-6. [Full Text]  

This article has been cited by other articles:

				K. E. Kramer, P. J. Rice, B. P. Horgan, J. L. Rittenhouse, and K. W. King Pesticide Transport with Runoff from Turf: Observations Compared with TurfPQ Model Simulations J. Environ. Qual., October 29, 2009; 38(6): 2402 - 2411. [Abstract] [Full Text] [PDF]

				K. W. King, J. C. Balogh, K. L. Hughes, and R. D. Harmel Nutrient Load Generated by Storm Event Runoff from a Golf Course Watershed J. Environ. Qual., May 25, 2007; 36(4): 1021 - 1030. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Related articles in JEQ Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Schwartz, L. Articles by Shuman, L. M. Search for Related Content PubMed PubMed Citation Articles by Schwartz, L. Articles by Shuman, L. M. Agricola Articles by Schwartz, L. Articles by Shuman, L. M. Related Collections Watershed and Landscape Processes Surface Water Quality Nutrients Solute Transport Models Runoff