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

This Article Abstract 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 ISI 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 ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Walden, R. R. Articles by Haith, D. A. Search for Related Content PubMed PubMed Citation Articles by Walden, R. R. Articles by Haith, D. A. GeoRef GeoRef Citation Agricola Articles by Walden, R. R. Articles by Haith, D. A. Related Collections Evapotranspiration Models Turfgrass Management Turfgrass Pesticides Air Pollution

SHORT COMMUNICATION

Estimating Turf Pesticide Volatilization from Simple Evapotranspiration Models

^a Blue: Land, Water, Infrastructure, 115 E. First St., Clayton, NC 27520
^b Dep. of Biological and Environmental Engineering, Riley-Robb Hall, Cornell Univ., Ithaca, NY 14853

^* Corresponding author (dah13{at}cornell.edu)

Received for publication May 20, 2002.

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
A previously developed model by Haith et al. (2002) related pesticide volatilization from turf to evapotranspiration (ET) by scaling factors determined from vapor pressures and heats of vaporization. Although the model provided volatilization estimates that compared well with field measurements, it relied on the Penman ET equation, requiring hourly temperature, wind speed, and solar radiation data, none of which are routinely available at field sites. The current study determined that the volatilization model works equally well with a simpler ET equation requiring only daily temperatures. Three daily temperature-based ET models were evaluated as vehicles for estimating pesticide volatilization from turf: Hamon, Hargreaves–Samani, and a modified Priestley–Taylor. When compared with field volatilization measurements for eight pesticides, volatilization estimates produced from the Hargreaves–Samani model most closely approximated both the field observations and the previous estimates based on the more data-intensive Penman model. Mean estimated volatilization exceeded mean observations by 15% and the coefficient of variation (R²) between estimates and observations was 0.65. The comparable values based on Penman ET were 17% and 0.63, respectively.

Abbreviations: ET, evapotranspiration

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
FIELD EXPERIMENTS have measured significant quantities of volatilized pesticides following applications to turf (Cooper et al., 1990; Murphy et al., 1996a,b; Taylor et al., 1977; Turner et al., 1977). Comparisons of possible human exposure with chronic reference doses have indicated that some of these vaporized chemicals may produce health hazards (Haith et al., 2000; Murphy et al., 1996a). Most volatilization occurs within a few days of application and is greatest at midday when temperatures and solar radiation are the highest. Volatilization varies with pesticide properties and environmental conditions, however, and it is difficult to generalize experimental results to the range of chemicals and conditions found in turf systems. Alternatively, mathematical models could be used to estimate volatilization of turf pesticides. Models are plausible tools if they are both credible and practical. Credibility is established by comparisons of model results with field measurements; practicality is largely determined by the extent to which model input parameters can be inferred from readily available data.

Pesticide volatilization models are often based on equilibrium partitioning of the chemical into solid, liquid, and gaseous phases in the soil environment. However, turf pesticides are more typically volatilized directly from vegetation, and it is difficult to extend the equilibrium approaches to turf due to the uncertainties in identifying the appropriate gas, solid, and liquid components of the turf foliage and thatch. A more promising approach is based on the similarities between pesticide vaporization and water evaporation. Haith et al. (2002) developed a model for pesticide volatilization based on turfgrass evapotranspiration (ET). Pesticide vaporization was related to ET by scaling factors determined from vapor pressures and heats of vaporization. First-order degradation of the pesticide on the turf foliage and thatch was also assumed. The ET was computed from the Penman equation using hourly temperature, solar radiation, and windspeed data. The model was tested by comparisons of predictions with measurements of volatilization for eight pesticides over 3 to 7 d in 11 field experiments. Measurements for two of the pesticides were used for model calibration; data from the remaining six pesticides were used to validate model predictions. Predicted mean losses for the six validation pesticides exceeded observations by 20%, and the model explained 67% of the observed variation in volatilization fluxes.

The Haith et al. (2002) model appears to be a credible means of assessing pesticide volatilization from turf, but its reliance on the Penman ET equation limits its practicality. Solar radiation and windspeed data are available for relatively few sites, as are hourly temperature records. Simpler ET equations are available that require only daily temperatures (minimum and maximum values or daily means). Although these temperature-based models are generally considered less accurate than the Penman equation, they are routinely used in irrigation management when more extensive weather data is lacking.

This paper describes an evaluation of three daily temperature-based ET equations as vehicles for estimating turf pesticide volatilization. Each of the equations was used to provide ET values for the Haith et al. (2002) volatilization model. Resulting volatilization fluxes were compared with field measurements and also with fluxes obtained previously based on the Penman model.

	Materials and Methods

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Volatilization Model
The volatilization model described by Haith et al. (2002) is:

[1]

in which V_t is the pesticide vaporized from surface vegetation (grass and thatch) during time period t (g ha^-1), k is a volatilization constant (mm^-1), R_t is the relative volatility of the chemical and water during period t (mm), and C_t is the pesticide available for volatilization on the vegetation at the beginning of period t (g ha^-1). The relative volatility is given by:

[2]

where ET_t is evapotranspiration during period t (mm); p_sct and p_swt are the saturated vapor pressures of the chemical and water, respectively, during period t (kPa); and _ct and _wt are the latent heats of vaporization of the chemical and water, respectively, during period t (J g^-1).

Equation [2] adjusts the water evapotranspiration for differences in chemical and water saturated vapor pressures and energy required for vaporization. Methods for computing vapor pressures and heats of vaporization as functions of temperature are described in Haith et al. (2002).

Water heat of vaporization and vapor pressure is given by Jensen et al. (1990):

[3]

[4]

in which T_t = air temperature during period t (°C). Pesticide vapor pressures and heats of vaporization are determined from Grain (1982), as described in Haith et al. (2002):

[5]

[6]

In these equations, p_sc0 = vapor pressure (kPa) at absolute temperature T_a0 (K), _c0 = latent heat of vaporization (J g^-1) at T_a0, M = molecular weight of the chemical, R = gas constant (8.32 J mol^-1 - K), Z_b = compressibility factor at boiling point (dimensionless), T_at = absolute temperature during period t (K), and m is equal to 0.19 and 0.8 for liquid and solid chemicals, respectively. Based on examples given in Grain (1982), Z_b = 0.97 and K_f = 1.06. The chemical latent heat of vaporization at time t, _ct, is also estimated from Eq. [6], with T_at substituted for T_a0.

Decay of the pesticide through bio- or photochemical degradation is assumed first-order, and the overall pesticide mass balance is:

[7]

in which is the degradation rate of the pesticide on the turf surface (time^-1). Other pesticide losses such as runoff, leaching, or removal of clippings are neglected.

Evapotranspiration Models
The ET variable in Eq. [2] can be calculated from alternative models for potential ET. The original application in Haith et al. (2002) used the Penman equation with hourly weather data, as described in Jensen et al. (1990). In the current study, three different daily temperature-based potential ET methods, the Hamon, Hargreaves–Samani, and Priestley–Taylor equations, are used for this same purpose. The only weather data required for the first two models are daily air temperatures, while the Priestley–Taylor equation requires both daily temperatures and solar radiation. We modified the equation by approximating solar radiation from mean monthly ratios of actual to maximum sunshine hours, so all three models could be implemented with only daily temperatures.

The Hamon (1961) equation is:

[8]

in which D_t is the number of daylight hours during day t and T_t here refers to the mean air temperature during day t (°C). The daylight hours variable can be determined by:

[9]

where _s is the is the sunset hour angle (radians), calculated from:

[10]

with equal to latitude (radians), and is the declination (radians) given by:

[11]

and DOY is the day of the year (1–365) (Jensen et al., 1990).

The second model is the Hargreaves–Samani equation as described by Jensen et al. (1990):

[12]

In this equation, T_Dt is the difference between the daily maximum and minimum temperatures on day t (°C) and R_At is the extraterrestrial radiation on day t (kJ m^-2) as determined from the following:

[13]

[14]

where G_sc is the solar constant (82 kJ m^-2 min^-1) and d_r is the relative distance of the earth from the sun.

The last model used is the Priestley–Taylor equation (Jensen et al., 1990):

[15]

In this equation, _t is the slope of the saturation vapor pressure curve at T_t (kPa °C^-1), _t is the psychometric constant at T_t (kPa °C^-1), R_nt is the net radiant energy available at the surface during day t (kJ m^-2), and G_t is the net sensible heat flux from the surface to soil during day t (kJ m^-2). Parameters for Eq. [15] are given by Jensen et al. (1990) as:

[16]

[17]

[18]

In these equations, P is the mean atmospheric pressure at the site (kPa) and EL is the elevation (m).

The Priestley–Taylor model, as given by Eq. [15], is essentially the Penman model without vapor pressure and windspeed terms. Although it requires both temperature and solar radiation data, the latter can be approximated from extraterrestrial radiation. Also, as with the Penman equation, the sensible heat flux term is generally assumed to be zero. The net radiant energy (R_nt) can be estimated from the following regression equation developed for grass in Minnesota (Jensen et al., 1990):

[19]

where R_st (kJ m^-2) is incoming solar radiation, determined from extraterrestrial radiation as:

[20]

where n/N is the ratio between actual measured bright sunshine hours and maximum possible sunshine hours (Jensen et al., 1990). With the substitution of mean monthly values for n/N, the only weather data required are daily air temperatures.

Model Testing
Model testing was based on the same experimental data set used in Haith et al. (2002). The testing sites were 0.2-ha plots with Hadley silt loam (coarse-silty, mixed, superactive, nonacid, mesic Typic Udifluvent) at the University of Massachusetts Turfgrass Research Center in South Deerfield, MA. The turf was well-established creeping bentgrass (Agrostis palustris L.) maintained at 13 mm with thatch thickness from 10 to 15 mm. Data obtained from these experiments included the measured concentrations of volatile residues in the air following applications of eight pesticides in 11 experiments conducted in the growing seasons of 1995, 1996, and 1997 (Table 1). Ethoprop (O-ethyl S,S-dipropyl phosphorodithioate) was applied in seven of the experiments, isofenphos [1-methylethyl 2-((ethoxy((1-methylethyl)amino)phosphinothioyl)oxy) benzoate] in six, and bendiocarb (2,2-dimethyl-1,3-benzoldioxol-4-yl methylcarbamate), carbaryl (1-naphthyl-N-methylcarbamate), chlorpyrifos [O,O-diethyl O-(3,5,6-trichloro-2-pyridyl) phosphorothioate], diazinon [O,O-diethyl O-(2-isopropyl-6-methyl-4-pyrimidinyl)phosphorothioate], isazofos [O,O-diethyl O-(5-chloro-1-(1-methylethyl)-1H-1,2,4-triazol-3-yl) phosphorothioate], and trichlorfon [dimethyl (2,2,2-trichloro-1-hydroxyethyl)phosphonate] were each applied in four of the experiments. Volatile residues were collected during selected sampling intervals of 1 to 4 h for up to 7 d following application. The theoretical profile shape method (Murphy et al., 1996a,b; Wilson et al., 1982) was used to estimate volatilization mass flux for each interval. The sampling program is described in more detail in Haith et al. (2002).

View this table: [in this window] [in a new window]   Table 2. Pesticide properties.

Equations [1] through [7] were used to predict the volatilization mass flux (V_t) for each pesticide. In the original application with the Penman equation, ET, saturated vapor pressures, and latent heats of vaporization were all determined for each 1- to 4-h sampling interval, based on the measured hourly temperatures, solar radiation, and wind speeds during that interval. In the current study, the equations were used with a daily time step, and ET, vapor pressures, and heats of vaporization were determined from daily temperatures. Thus, the three daily ET models and the associated volatilization model provide ET and volatilization estimates for the 24-h period containing the sampling intervals, whereas the comparable estimates from the Penman-based computations are for the actual sampling intervals. It should thus be expected that volatilization estimates from the daily models would exceed both measured fluxes and those estimated from Penman ET. However, at least for the first 3 d of each experiment, sampling intervals cover most of the daylight hours, and should correspond to the periods of significant ET and volatilization. After Day 3 of each experiment, volatile residues were only collected for 4 h each day. However, the rates of volatilization on these later days of the experiments were generally so small that lack of additional sampling intervals would have little effect on the total volatilization measured in the experiment.

	Results and Discussion

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 
Evapotranspiration
The ET estimates from the three daily models are compared with Penman estimates for the first three days of each experiment in Table 3. There is clearly a great deal of variation among the four estimates in any one experiment. Furthermore, the differences are very inconsistent between experiments. For example, in Experiment 96-2, the Hargreaves–Samani value is relatively close to the Penman, but the other two models produce much lower values. In Experiment 97-2, the Hamon estimate is closest to the Penman value and the Hargreaves–Samani estimate is much larger.

One of the reasons for the relatively poor performance of the Priestley–Taylor model may be the use of monthly mean values for solar radiation, as determined by Eq. [19] and [20]. The use of actual daily solar radiation data, or other approaches for estimating monthly means, might produce ET estimates closer to the Penman values.

Pesticide Volatilization
Observed and modeled volatilization mass fluxes are compared for each of the eight pesticides in Table 4. The means and coefficients of determination for the Penman results differ somewhat from those reported in Haith et al. (2002), because the latter were based only on the six pesticides used for model validation. For all pesticides, fluxes based on the Hargreaves–Samani ET model are closer to the Penman results than fluxes based on the other two daily models. More critically, the Hargreaves–Samani fluxes more closely match the measured fluxes for all pesticides except ethoprop. The Priestley–Taylor and Hamon ET values produced more accurate ethoprop estimates than the other ET models, including the Penman. For the Priestley–Taylor, this resulted in an overall mean volatilization flux that was within 14% of measurements, virtually identical to the 15% difference seen in the Hargreaves–Samani results. Nevertheless, the latter must be considered superior because it was more accurate for seven of the eight chemicals and also produced a conservative overprediction of mean losses, a desirable trait in a model used for assessment of health effects.

The similarity in the Penman and Hargreaves–Samani volatilization fluxes is somewhat surprising given the substantial differences in ET values produced by the two models. Mean ET from the Hargreaves–Samani model exceeds the Penman mean by 23% (Table 3), and since the volatilization model is a linear function of ET, as indicated by Eq. [2], it would appear that volatilization fluxes should show similar differences. However, the vapor pressures used in Eq. [2] are nonlinear, increasing functions of temperature, as indicated by Eq. [5]. With the daily models, vapor pressures are based on the mean temperature for the day, whereas with the hourly data used with the Penman model, vapor pressures are computed from the actual temperatures during the sampling period. As a result, calculated vapor pressures and volatilization based on hourly temperatures would typically be higher than comparable values obtained from daily temperatures. It appears that the somewhat higher ET estimates compensate for the lower vapor pressures that result from use of mean daily temperatures.

Effects of Model Recalibration
The modeled volatilization fluxes in Table 4 were all calculated using the volatilization constants (k) from Haith et al. (2002) that were calibrated to fit the Penman results. It seems reasonable that better fits between the fluxes determined from the daily ET models and the observed fluxes could be obtained if the volatilization fluxes were recalibrated to these models. To explore this possibility, new volatilization constants were calculated for the Hargreaves–Samani and Priestley–Taylor models. As with the previous Penman calibrations, isazofos and trichlorfon were used as the calibration pesticides for their groups because their measured volatilization losses placed them in the middle of their respective groups. The new constants that were obtained for Group 1 and 2 chemicals, respectively, were k = 155 and 460 mm^-1 for the Hargreaves–Samani model and k = 215 and 620 mm^-1 for the Priestley–Taylor model.

The mean volatilization fluxes for each pesticide based on the new calibrations are shown in Table 5. Although the estimated fluxes improved for some of the pesticides, most notably chlorpyrifos, the coefficients of determination were approximately the same, and overall means were somewhat worse. In general, there appears to be little advantage in recalibrating the volatilization model to the individual ET model, and the original volatilization constants are valid for these daily ET models.

	Conclusions

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

Three daily temperature-based ET models were evaluated as vehicles for estimating pesticide volatilization from turf: Hamon, Hargreaves–Samani, and a modified Priestley–Taylor. When compared with field volatilization measurements for eight pesticides, volatilization estimates produced from the Hargreaves–Samani model most closely approximated both the field observations and the previous estimates based on the more data-intensive Penman model. Mean estimated volatilization exceeded mean observations by 15%. The coefficient of variation (R²) between estimates and observations was 0.65. The comparable values based on Penman ET were 17% and 0.63, respectively. The remaining two ET models produced fluxes that underestimated mean volatilization by 14 to 21%.

The credibility or accuracy of the Haith et al. (2002) ET-based pesticide volatilization model was not reduced when Penman ET estimates were replaced by estimates from the simpler, Hargreaves–Samani equation. This significantly increases the practicality of the volatilization model, since it reduces the required weather inputs to daily temperatures, which should be available at most field sites.

	ACKNOWLEDGMENTS

 
Research described in this paper was supported, in part, by Green Section Research, U.S. Golf Association.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Conclusions REFERENCES

 

• Cooper, R.J., J.J. Jenkins, and A.S. Curtis. 1990. Pendimethalin volatility following application to turfgrass. J. Environ. Qual. 19:508–513.[Abstract/Free Full Text]
• Grain, C.F. 1982. Vapor pressure. p. 14-1 to 14-20. In W.J. Lyman, W.F. Reehl, and D.H. Rosenblatt (ed.) Handbook of chemical property estimation methods. McGraw-Hill, New York.
• Haith, D.A., C.J. DiSante, G.R. Roy, and J.M. Clark. 2000. Modeling approaches for assessment of exposure from volatilized pesticides applied to turf. p. 255–267. In J.M. Clark and M.P. Kenna (ed.) Fate and management of turfgrass chemicals. ACS Symp. Ser. 743. Am. Chem. Soc., Washington, DC.
• Haith, D.A., P.C. Lee, J.M. Clark, G.R. Roy, M.J. Imboden, and R.R. Walden. 2002. Modeling pesticide volatilization from turf. J. Environ. Qual. 31:724–729.[Abstract/Free Full Text]
• Hamon, W.R. 1961. Estimating potential evapotranspiration. Proc. Am. Soc. Civ. Eng. J. Hydraul. Div. 87:107–120.
• Jensen, M.E., R.D. Burman, and R.G. Allen (ed.) 1990. Evapotranspiration and irrigation water requirements. ASCE Manual no. 70. Am. Soc. of Civ. Eng., New York.
• National Climatic Data Center. 2002. Sunshine—Average percent (%) possible. Available online at http://lwf.ncdc.noaa.gov/oa/climate/online/ccd/avgsun.html (verified 24 Jan. 2003). NCDC, Asheville, NC.
• Northeast Regional Climate Center. 2002. Climod database. Available online at http://climod.nrcc.cornell.edu/ (verified 24 Jan. 2003). NRCC, Cornell Univ., Ithaca, NY.
• Murphy, K.C., R.J. Cooper, and J.M. Clark. 1996a. Volatile and dislodgeable residues following trichlorfon and isazofos application to turfgrass and implications for human exposure. Crop Sci. 36:1446–1454.[Abstract/Free Full Text]
• Murphy, K.C., R.J. Cooper, and J.M. Clark. 1996b. Volatile and dislodgeable residues following triadimefon and MCPP application to turfgrass and implications for human exposure. Crop Sci. 36:1455–1461.[Abstract/Free Full Text]
• Taylor, A.W., D.E. Glotfelty, B.C. Turner, R.E. Silver, H.P. Freeman, and A. Weiss. 1977. Volatilization of dieldrin and heptachlor residues from field vegetation. J. Agric. Food Chem. 25:542–548.[ISI][Medline]
• Turner, B.C., D.E. Glotfelty, and A.W. Taylor. 1977. Photodieldrin formation and volatilization from grass. J. Agric. Food Chem. 25:548–550.[ISI][Medline]
• Wilson, J.D., G.W. Thurtell, G.E. Kidd, and E.G. Beauchamp. 1982. Estimation of the rate of gaseous mass transfer from a surface source plot to the atmosphere. Atmos. Environ. 16:1861–1867.

Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2003 32: 745-750. [Full Text]  

This article has been cited by other articles:

				A. Wolters, M. Klein, and H. Vereecken An Improved Description of Pesticide Volatilization: Refinement of the Pesticide Leaching Model (PELMO) J. Environ. Qual., September 1, 2004; 33(5): 1629 - 1637. [Abstract] [Full Text] [PDF]

This Article Abstract 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 ISI 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 ISI Web of Science (2) Citing Articles via Google Scholar Google Scholar Articles by Walden, R. R. Articles by Haith, D. A. Search for Related Content PubMed PubMed Citation Articles by Walden, R. R. Articles by Haith, D. A. GeoRef GeoRef Citation Agricola Articles by Walden, R. R. Articles by Haith, D. A. Related Collections Evapotranspiration Models Turfgrass Management Turfgrass Pesticides Air Pollution