Model Development Default Model Functions included in the model are: Hydrologic functions: Depth to...
1
Model Development Model Development Default Model Default Model •Functions included in the model are: • Hydrologic functions: Depth to Free Water, Flooding Frequency, Index of Surface Runoff (ISRO) (Fig. 2), and Slope Length Erodibility (LSK wf ) • Dissipation functions: Biological Degradation, Adsorption, Hydrolysis, Volatilization •Based on previous research in claypan watersheds (Lerch et al., 1995; Blanchard and Donald, 1997; Donald et al., 1998), it was evident that the default model over-estimated leaching risk and underestimated risk of surface runoff and particle adsorbed runoff for claypan soils (Fig. 3). Restrictive Layer Criteria Restrictive Layer Criteria • The default model modified risk by increasing the ISRO weight based on the minimum K sat within a soil profile, but this was not sufficient to accurately characterize risk for claypan watersheds. To address this, it was necessary to develop a restrictive soil layer criteria to be applied in watersheds, such as the claypan watershed included in this study, with hydrologically restrictive layers. •In claypan watersheds, argillic horizons represent the most hydrologically restrictive layer and control the movement of water. The criteria was developed based on the presence of argillic horizons (B t ) in the SSURGO profile descriptions. The amount of clay and the thickness of the clay layer are both important factors related to the intensity of a restrictive clay layer. Therefore, a B t intensity (B ti ) factor was developed. • B ti = Σ (thickness of B t horizon(s)) x %clay content = cm of clay. •Soils with B ti greater than or equal to 35 were considered to have restrictive clay layers. • The B ti was then used as a criterion for applying and/or modifying several of the hydrologic functions. • Free water function - leaching risk was decreased for soils with a restrictive layer (B ti >=35) above the depth to the water table. • ISRO function - for soils with B >=35, ISRO was increased based on the available water holding capacity (AWC) Assessment of Herbicide Transport in Central Missouri Assessment of Herbicide Transport in Central Missouri Watersheds Watersheds Using a Process-Based Index Model Using a Process-Based Index Model C.D. Willett 1 , R.N. Lerch 2 , M. Milner 3 , M. L. Bernards 3 , and P. J. Shea 4 1 Department of Soil, Environmental and Atmospheric Sciences, University of Missouri; 2 USDA-Agricultural Research Service, Cropping Systems and Water Quality Research Unit, Columbia, MO; 3 Department of Agronomy and Horticulture, University of Nebraska-Lincoln; 4 School of Natural Resources, University of Nebraska-Lincoln. Abstract Abstract Identification of areas at high risk for herbicide loss is a necessary requirement for targeting best management practices (BMPs) to the most vulnerable areas within fields or watersheds. To work toward that goal, a process-based index model has been developed to assess relative landscape vulnerability to hydrologic losses of herbicides. The model applies mathematical functions to assign scores (i.e., degree of risk) on the basis of herbicide, soil, and hydrologic properties relevant to the environmental fate of herbicides. The model uses the NRCS county SSURGO (Soil Survey Geographic) database as the input source of soils and hydrologic data. The risk of herbicide transport is considered for three hydrologic pathways, or scenarios: leaching, surface runoff, and particle adsorbed runoff. The model is used to compute the relative risk of herbicide transport for a given herbicide for each hydrologic pathway, providing the spatial and temporal risk of transport. The model was applied to the Youngs Creek watershed in the Central Claypan Region of Missouri, a watershed with known herbicide contamination problems. The risk of herbicide loss for each hydrologic pathway was determined for each of four corn herbicides (atrazine, metolachlor, isoxaflutole, and glyphosate). By computing area-weighted scores at the watershed scale, a score for each herbicide and hydrologic pathway scenario combination was determined. The area- weighted scores were used to evaluate the potential spatial and temporal risk of applying these herbicides in the Youngs Creek watershed. The information provided by the model can be used to make recommendations regarding the choice of herbicides that will minimize the risk of hydrologic transport for this watershed. Conclusions Conclusions •Development of restrictive clay layer criteria was essential for accurately assessing hydrologic risk of herbicide transport in a claypan watershed. •The model is sensitive to landscape position and herbicide properties. •The index model approach allows for watershed-scale quantification of herbicide risk to transport using time-integrated, area-weighted soil mapping unit scores. •Maximum sensitivity requires that the number of comparisons over time or herbicide choices be limited (i.e. pair-wise comparisons of herbicides, limit time scale to 60 days) •The developed model can be used to target management practices to the most vulnerable areas within watersheds and as a decision support tool for making environmentally sound herbicide choices. References References Ahrens, W.H., ed. 1994. Herbicide Handbook 7th ed. Champaign, IL: Weed Science Society of America. 352 p. Blandard, P.E., and W.W. Donald. 1997. Herbicide contamination of groundwater beneath claypan soils in north- central Missouri. J. Environ. Qual. 26:1612-1621. Bresnaham, G.A., W.C. Koskinen, A.G. Dexter, and L. Cox. 2004. Sorption-desorption of ‘aged’ isoxaflutole and diketonitrile degradate in soil. Weed Research 44:397-403 Kazemi, H.V., S.H. Anderson, K.W. Goyne, and C.J. Gantzer. 2008. Atrazine and alachlor transport in claypan soils as influenced by differential antecedent soil water content. J. Environ. Qual. 37:1599-1607. Koskinen, W.C., T.E.Ochsner, B.M. Stephens, and R.S. Kookana. 2006. Sorption of isoxaflutol diketonitrile degradate (DKN) and dicamba in unsaturated soil. J. Environ. Sci. Health Part B 41:1701-1083. Lerch, R.N., W.W. Donald, Y. Li, and E.E. Alberts. 1995. Hydroxylated atrazine degradation products in a small Missouri stream. Environ. Sci. Technol. 29:2759-2768. Lerch, R.N., and P.E. Blanchard. 2003. Watershed vulnerability to herbicide transport in northern Missouri and southern Iowa streams. Environ. Sci. Technol. 37:5518-5527. Lin, C.H., R.N. Lerch, H.E. Garrett, W.G. Johnson, D. Jordan, and M.F. George. 2003 The effect of five forage species on transport and transformation of atrazine and isoxaflutole (Balance) in lysimeter leachate. J. Environ. Qual. 32:1992-2000. Geisy, J.P., S. Dobson, and K.R. Solomon. 2000. Ecotoxicological risk assessment for Roundup herbicide. Reviews Environ. Contam. Toxicol. 167:35-120. Montgomery, J.H. 1993. Agrochemicals Desk Reference. Chelsea, MI: Lewis Publishers. 625 p. Seybold, C.A and W. Mersie. 1996. Adsorption and desorption of atrazine, deethylatrazine, deisoproplyatrazine, hydroxyatrazine, and metolachlor in two soils from Virginia. J. Environ. Qual. 25:1179-1185. Soil Survey Staff, 2005. National collection of soil survey geographic data. SSURGO Version 2.1 via Soil Data Warehouse. U.S. Department of Agriculture, Natural Resource Conservation Service, National Soil Survey Center, Lincoln, NE. Digital soil survey area maps and attribute tables, electronic files as accessed through Hydrologic Pathways Hydrologic Pathways • Relative risk between hydrologic pathways was consistent with field data: ARO > SRO > Leaching (Figs. 4-7; Table 1) (Lerch, 1995; Blanchard and Donald, 1997; Lerch and Blanchard, 2003). •High risk for herbicide transport by SRO and ARO is due to the combination of the restrictive clay layer and highly erodible surface soils. Landscape Sensitivity Landscape Sensitivity • The model is sensitive to differences in soil properties across the landscape •Alluvial soils have a high risk of leaching, upland soils have a high risk for transport by SRO and ARO. Herbicide Comparisons Herbicide Comparisons • Model correctly discerned differences among herbicides with a wide range of properties •The model was sensitive to differences in K d and half-life •The herbicides with the highest sorption intensity (glyphosate and metolachlor) showed the highest risk for transport by ARO and the lowest risk for leaching. Conversely, atrazine and DKN showed highest risks for loss by leaching and SRO. •DKN and metolachlor showed the lowest risks for all pathways by Day 30, reflecting their shorter half-lives compared to atrazine and glyphosate. Watershed Scores Watershed Scores • The time-integrated watershed scores can be used for decision-making regarding herbicide choice(s) that would result in the least contamination of water resources (Table 1). Materials and Methods Materials and Methods Site Description Site Description Youngs Creek Watershed •227.1 km 2 agricultural watershed in north-central Missouri (Fig. 1) •Watershed boundaries within Audrain, Boone, and Monroe Counties •Part of the Central Claypan Region (MLRA 113) •Representative of an intensely row cropped region •Gently sloping topography, typical slope 1-4% •Soils characterized by restrictive clay layers and high runoff and erosion potential Objective Objective Evaluate the risk of herbicide loss through leaching, surface runoff, and particle adsorbed runoff using components of landscape, soil, and herbicide properties for the purpose of targeting conservation practices at the watershed scale. Model Inputs Model Inputs • Data obtained from the SSURGO database (Soil Survey Staff, 2005) • Soils: profile descriptions, organic matter, clay content, pH, erodibility factor (K wf ) • Hydrology: saturated hydraulic conductivity (K sat ), depth to water table, available water content (AWC), slope (S), slope length (L) • Herbicide inputs- sorption coefficient (K d ) used to compute soil-specific K oc values, half-life (t 1/2 ), and vapor pressure Herbicide Selection Herbicide Selection •Based on widespread use in the Midwest •Represents four separate herbicide classes •Atrazine: K d = 1.44; t 1/2 = 36 d (Kazemi et al., 2008); vp = 2.9 X 10 -7 mm Hg (Ahrens, 1994) • Diketonitrile (DKN) metabolite of Isoxaflutole: K d = 0.77 (Bresnaham et al., 2004, Koskinen et al., 2006); t 1/2 = 12 d (Lin et al., 2003) •Glyphosate: K d = 55 (Montgomery, 1993); t 1/2 = 40 d (Geisy et al., 2000) •Metolachlor: K d = 1.83 (Seybold and Mersie, 1996, Weber et al., 2000); t 1/2 = 27 d (Montgomery, 1993); vp = 3.1 X 10 -5 mm Hg (Montgomery, 1993) Assessing Risk Assessing Risk A series of equations for soil, hydrologic, and time-dependent dissipation functions were developed for each hydrologic pathway: leaching, surface runoff (SRO), and particle adsorbed runoff (ARO). Raw scores (R S ) from each function were summed for each soil-herbicide-time combination. The R S were then transformed into standardized scores (SS) with a scale of 0 (lowest risk) to 10 (highest risk) that allow for comparisons across all times, herbicides, and soils. The SS were then combined with the spatial distribution of each mapping unit to develop maps for each soil-herbicide-time combination using ArcView 3.3. An area-weighted score was computed for each herbicide and time combination by multiplying the SS for a given soil mapping unit by the fractional area of each mapping unit within the watershed, then summing the scores. The area-weighted scores were then plotted against time (Fig. 7), and a time-integrated watershed score (W s ) was then determined by integrating the area under each curve for a given herbicide and hydrologic pathway (Table 1). Results and Discussion Results and Discussion Figure 1. Location of Youngs Creek Watershed Figure 1. Location of Youngs Creek Watershed Figure 2. Example of the function Figure 2. Example of the function for computing ISRO. R for computing ISRO. R S = (2 + = (2 + (1.037*e (1.037*e -0.7934*S -0.7934*S + 2.6819* + 2.6819* e-0.0617 e-0.0617 *S + *S + 0.435 ln(minKsat)) + AWC Penalty 0.435 ln(minKsat)) + AWC Penalty Figure 3. Application of the default model functions to the claypan soils of Figure 3. Application of the default model functions to the claypan soils of Audrain County, MO showed highest risk for herbicide transport by leaching Audrain County, MO showed highest risk for herbicide transport by leaching and lowest risk for SRO. This finding was in opposition to field studies in and lowest risk for SRO. This finding was in opposition to field studies in claypan watersheds. claypan watersheds. Figure 6. Particle Adsorbed Figure 6. Particle Adsorbed Runoff Risk Runoff Risk Day 1 Day 30 Day 7 Day 100 Day 180 Particle Adsorbed Runoff Risk at Day 0 Metolachlor Glyphosate Atrazine DKN Figure 4. Leaching Figure 4. Leaching Risk Risk Day 1 Day 30 Day 7 Day 100 Day 180 Leaching Risk at Day 0 Metolachlor Glyphosate Atrazine DKN Model Limitations Model Limitations • Risk at Day 0 was over-estimated for leaching. • Time-dependent functions (degradation, adsorption, hydrolysis, and volatilization) under-estimated dissipation. •All hydrologic pathways should converge to very low risk by Day 180. For the compounds included in this study, at least 4 half-lives had occurred by Day 180. • Model sensitivity will be greatest when the number of comparisons are limited. •Temporal comparisons should be limited to the critical transport period of ~60 days after application. •Herbicide comparisons should be limited to two compounds. Results and Discussion (con’t.) Results and Discussion (con’t.) Herbicide Leaching Surface Runoff Particle Adsorbed Runoff Mean --------------------------- Score•Days ----------------------------- Atrazine 634† 981 1128 914 DKN 617 964 1022 868 Glyphosat e 588 934 1207 910 Metolachl 587 933 1114 878 Table 1. Time-integrated watershed scores (W Table 1. Time-integrated watershed scores (W s ) for each herbicide and ) for each herbicide and hydrologic pathway. hydrologic pathway. Figure 5. Surface Figure 5. Surface Runoff Risk Runoff Risk Day 1 Day 30 Day 7 Day 100 Day 180 Surface Runoff Risk at Day 0 Metolachlor Glyphosate Atrazine DKN Slope (S) 0 10 20 30 40 50 ISR O Weight 0 1 2 3 4 5 6 7 8 -C ap at7 Points -U se upperm ostm in K satin profile -AW C Penalty notapplied to graph 100 m m sec -1 10 m m sec -1 1m msec -1 0.1m m sec -1 0.01m m sec -1 † The lower the W s, the lower the risk. Figure 7. Area-weighted risk scores over time for atrazine, DKN, Figure 7. Area-weighted risk scores over time for atrazine, DKN, glyphosate, and metolachlor for each hydrologic pathway. This data was glyphosate, and metolachlor for each hydrologic pathway. This data was used to compute the watershed scores (W used to compute the watershed scores (W s ) in Table 1. ) in Table 1. Leaching Leaching SRO SRO ARO ARO
-
Upload
milo-wilkinson -
Category
Documents
-
view
215 -
download
0
Transcript of Model Development Default Model Functions included in the model are: Hydrologic functions: Depth to...
Model DevelopmentModel DevelopmentDefault ModelDefault Model•Functions included in the model are:
•Hydrologic functions: Depth to Free Water, Flooding Frequency, Index of Surface Runoff (ISRO) (Fig. 2), and Slope Length Erodibility (LSKwf)
•Dissipation functions: Biological Degradation, Adsorption, Hydrolysis, Volatilization•Based on previous research in claypan watersheds (Lerch et al., 1995; Blanchard and Donald, 1997; Donald et al., 1998), it was evident that the default model over-estimated leaching risk and underestimated risk of surface runoff and particle adsorbed runoff for claypan soils (Fig. 3).
Restrictive Layer CriteriaRestrictive Layer Criteria•The default model modified risk by increasing the ISRO weight based on the minimum Ksat within a soil profile, but this was not sufficient to accurately characterize risk for claypan watersheds. To address this, it was necessary to develop a restrictive soil layer criteria to be applied in watersheds, such as the claypan watershed included in this study, with hydrologically restrictive layers.
•In claypan watersheds, argillic horizons represent the most hydrologically restrictive layer and control the movement of water. The criteria was developed based on the presence of argillic horizons (Bt) in the SSURGO profile descriptions. The amount of clay and the thickness of the clay layer are both important factors related to the intensity of a restrictive clay layer. Therefore, a Bt intensity (Bti) factor was developed.
•Bti= Σ (thickness of Bt horizon(s)) x %clay content = cm of clay.
•Soils with Bti greater than or equal to 35 were considered to have restrictive clay layers.
•The Bti was then used as a criterion for applying and/or modifying several of the hydrologic functions.
•Free water function - leaching risk was decreased for soils with a restrictive layer (Bti>=35) above the depth to the water table.
•ISRO function - for soils with Bti>=35, ISRO was increased based on the available water holding capacity (AWC) above the restrictive layer. The lower the AWC above the restrictive layer, the greater the risk of herbicide transport by SRO or ARO pathways.
•LS and LSKwf functions - soils with restrictive clay layers are more prone to runoff and erosion than soils with similar slope and slope length without a restrictive clay layer. Therefore, a Severe Runoff Penalty was developed that increased the calculated LS value for restrictive layer soils to reflect their increased risk of SRO and ARO.
Assessment of Herbicide Transport in Central Missouri Watersheds Assessment of Herbicide Transport in Central Missouri Watersheds Using a Process-Based Index ModelUsing a Process-Based Index Model
C.D. Willett1, R.N. Lerch2, M. Milner3, M. L. Bernards3, and P. J. Shea4
1Department of Soil, Environmental and Atmospheric Sciences, University of Missouri; 2USDA-Agricultural Research Service, Cropping Systems and Water Quality Research Unit, Columbia, MO; 3Department of Agronomy and Horticulture, University of Nebraska-Lincoln; 4School of Natural Resources, University of Nebraska-Lincoln.
AbstractAbstractIdentification of areas at high risk for herbicide loss is a necessary requirement for targeting best management practices (BMPs) to the most vulnerable areas within fields or watersheds. To work toward that goal, a process-based index model has been developed to assess relative landscape vulnerability to hydrologic losses of herbicides. The model applies mathematical functions to assign scores (i.e., degree of risk) on the basis of herbicide, soil, and hydrologic properties relevant to the environmental fate of herbicides. The model uses the NRCS county SSURGO (Soil Survey Geographic) database as the input source of soils and hydrologic data. The risk of herbicide transport is considered for three hydrologic pathways, or scenarios: leaching, surface runoff, and particle adsorbed runoff. The model is used to compute the relative risk of herbicide transport for a given herbicide for each hydrologic pathway, providing the spatial and temporal risk of transport. The model was applied to the Youngs Creek watershed in the Central Claypan Region of Missouri, a watershed with known herbicide contamination problems. The risk of herbicide loss for each hydrologic pathway was determined for each of four corn herbicides (atrazine, metolachlor, isoxaflutole, and glyphosate). By computing area-weighted scores at the watershed scale, a score for each herbicide and hydrologic pathway scenario combination was determined. The area-weighted scores were used to evaluate the potential spatial and temporal risk of applying these herbicides in the Youngs Creek watershed. The information provided by the model can be used to make recommendations regarding the choice of herbicides that will minimize the risk of hydrologic transport for this watershed.
ConclusionsConclusions•Development of restrictive clay layer criteria was essential for accurately assessing hydrologic risk of herbicide transport in a claypan watershed.
•The model is sensitive to landscape position and herbicide properties.•The index model approach allows for watershed-scale quantification of herbicide risk to transport using time-integrated, area-weighted soil mapping unit scores.
•Maximum sensitivity requires that the number of comparisons over time or herbicide choices be limited (i.e. pair-wise comparisons of herbicides, limit time scale to 60 days)
•The developed model can be used to target management practices to the most vulnerable areas within watersheds and as a decision support tool for making environmentally sound herbicide choices.ReferencesReferences
Ahrens, W.H., ed. 1994. Herbicide Handbook 7th ed. Champaign, IL: Weed Science Society of America. 352 p. Blandard, P.E., and W.W. Donald. 1997. Herbicide contamination of groundwater beneath claypan soils in north-central Missouri. J.
Environ. Qual. 26:1612-1621.Bresnaham, G.A., W.C. Koskinen, A.G. Dexter, and L. Cox. 2004. Sorption-desorption of ‘aged’ isoxaflutole and diketonitrile
degradate in soil. Weed Research 44:397-403Kazemi, H.V., S.H. Anderson, K.W. Goyne, and C.J. Gantzer. 2008. Atrazine and alachlor transport in claypan soils as influenced by
differential antecedent soil water content. J. Environ. Qual. 37:1599-1607. Koskinen, W.C., T.E.Ochsner, B.M. Stephens, and R.S. Kookana. 2006. Sorption of isoxaflutol diketonitrile degradate (DKN) and
dicamba in unsaturated soil. J. Environ. Sci. Health Part B 41:1701-1083.Lerch, R.N., W.W. Donald, Y. Li, and E.E. Alberts. 1995. Hydroxylated atrazine degradation products in a small Missouri stream.
Environ. Sci. Technol. 29:2759-2768.Lerch, R.N., and P.E. Blanchard. 2003. Watershed vulnerability to herbicide transport in northern Missouri and southern Iowa
streams. Environ. Sci. Technol. 37:5518-5527.Lin, C.H., R.N. Lerch, H.E. Garrett, W.G. Johnson, D. Jordan, and M.F. George. 2003 The effect of five forage species on transport
and transformation of atrazine and isoxaflutole (Balance) in lysimeter leachate. J. Environ. Qual. 32:1992-2000. Geisy, J.P., S. Dobson, and K.R. Solomon. 2000. Ecotoxicological risk assessment for Roundup herbicide. Reviews Environ. Contam.
Toxicol. 167:35-120.Montgomery, J.H. 1993. Agrochemicals Desk Reference. Chelsea, MI: Lewis Publishers. 625 p. Seybold, C.A and W. Mersie. 1996. Adsorption and desorption of atrazine, deethylatrazine, deisoproplyatrazine, hydroxyatrazine,
and metolachlor in two soils from Virginia. J. Environ. Qual. 25:1179-1185.Soil Survey Staff, 2005. National collection of soil survey geographic data. SSURGO Version 2.1 via Soil Data Warehouse. U.S.
Department of Agriculture, Natural Resource Conservation Service, National Soil Survey Center, Lincoln, NE. Digital soil survey area maps and attribute tables, electronic files as accessed through http://soils.usda.gov.
Weber, J.B., G.G. Wilkerson, H.M. Linker, J.W. Wilcut, R.B. Leidy, S. Senseman, W.W. Witt, M. Barrett, W.K. Vencill, D.R. Shaw, T.C. Mueller, D.K. Miller, B.J. Brecke, R.E. Talbert, and T.F. Peeper. 2000. A proposal to standardize soil/solution herbicide distribution coefficients. Weed Sci. 48:75-88.
Hydrologic PathwaysHydrologic Pathways•Relative risk between hydrologic pathways was consistent with field data: ARO > SRO > Leaching (Figs. 4-7; Table 1) (Lerch, 1995; Blanchard and Donald, 1997; Lerch and Blanchard, 2003).
•High risk for herbicide transport by SRO and ARO is due to the combination of the restrictive clay layer and highly erodible surface soils.
Landscape SensitivityLandscape Sensitivity•The model is sensitive to differences in soil properties across the landscape
•Alluvial soils have a high risk of leaching, upland soils have a high risk for transport by SRO and ARO.
Herbicide ComparisonsHerbicide Comparisons•Model correctly discerned differences among herbicides with a wide range of properties
•The model was sensitive to differences in Kd and half-life
•The herbicides with the highest sorption intensity (glyphosate and metolachlor) showed the highest risk for transport by ARO and the lowest risk for leaching. Conversely, atrazine and DKN showed highest risks for loss by leaching and SRO.
•DKN and metolachlor showed the lowest risks for all pathways by Day 30, reflecting their shorter half-lives compared to atrazine and glyphosate.
Watershed ScoresWatershed Scores•The time-integrated watershed scores can be used for decision-making regarding herbicide choice(s) that would result in the least contamination of water resources (Table 1).
•Based on the Ws in Table 1, metolachlor and DKN would be the best choices for minimizing the risks of transport by SRO and ARO. Of course, other considerations will influence the “best” herbicide choice(s), such as application rates, weed control spectrum and cost.
Materials and MethodsMaterials and Methods
Site DescriptionSite DescriptionYoungs Creek Watershed•227.1 km2 agricultural watershed in north-central Missouri (Fig. 1)
•Watershed boundaries within Audrain, Boone, and Monroe Counties
•Part of the Central Claypan Region (MLRA 113)•Representative of an intensely row cropped region•Gently sloping topography, typical slope 1-4%•Soils characterized by restrictive clay layers and high runoff and erosion potential
ObjectiveObjectiveEvaluate the risk of herbicide loss through leaching, surface runoff, and particle adsorbed runoff using components of landscape, soil, and herbicide properties for the purpose of targeting conservation practices at the watershed scale.
Model InputsModel Inputs•Data obtained from the SSURGO database (Soil Survey Staff, 2005)
•Soils: profile descriptions, organic matter, clay content, pH, erodibility factor (Kwf)
•Hydrology: saturated hydraulic conductivity (Ksat), depth to water table, available water content (AWC), slope (S), slope length (L)
•Herbicide inputs- sorption coefficient (Kd) used to compute soil-specific Koc values, half-life (t1/2), and vapor pressure
Herbicide SelectionHerbicide Selection•Based on widespread use in the Midwest•Represents four separate herbicide classes•Atrazine: Kd = 1.44; t1/2 = 36 d (Kazemi et al., 2008); vp = 2.9 X 10-7 mm Hg (Ahrens, 1994)
•Diketonitrile (DKN) metabolite of Isoxaflutole: Kd = 0.77 (Bresnaham et al., 2004, Koskinen et al., 2006); t1/2 = 12 d (Lin et al., 2003)
•Glyphosate: Kd = 55 (Montgomery, 1993); t1/2 = 40 d (Geisy et al., 2000)
•Metolachlor: Kd = 1.83 (Seybold and Mersie, 1996, Weber et al., 2000); t1/2 = 27 d (Montgomery, 1993); vp = 3.1 X 10-5 mm Hg (Montgomery, 1993)
Assessing RiskAssessing RiskA series of equations for soil, hydrologic, and time-dependent dissipation functions were developed for each hydrologic pathway: leaching, surface runoff (SRO), and particle adsorbed runoff (ARO). Raw scores (RS) from each function were summed for each soil-herbicide-time combination. The RS were then transformed into standardized scores (SS) with a scale of 0 (lowest risk) to 10 (highest risk) that allow for comparisons across all times, herbicides, and soils. The SS were then combined with the spatial distribution of each mapping unit to develop maps for each soil-herbicide-time combination using ArcView 3.3. An area-weighted score was computed for each herbicide and time combination by multiplying the SS for a given soil mapping unit by the fractional area of each mapping unit within the watershed, then summing the scores. The area-weighted scores were then plotted against time (Fig. 7), and a time-integrated watershed score (Ws) was then determined by integrating the area under each curve for a given herbicide and hydrologic pathway (Table 1).
Results and DiscussionResults and Discussion
Figure 1. Location of Youngs Creek WatershedFigure 1. Location of Youngs Creek Watershed
Figure 2. Example of the function for Figure 2. Example of the function for computing ISRO. Rcomputing ISRO. RSS = (2 + (1.037*e = (2 + (1.037*e--
0.7934*S0.7934*S + 2.6819* + 2.6819*e-0.0617e-0.0617*S + 0.435 *S + 0.435 ln(minKsat)) + AWC Penaltyln(minKsat)) + AWC Penalty
Figure 3. Application of the default model functions to the claypan soils of Figure 3. Application of the default model functions to the claypan soils of Audrain County, MO showed highest risk for herbicide transport by leaching and Audrain County, MO showed highest risk for herbicide transport by leaching and lowest risk for SRO. This finding was in opposition to field studies in claypan lowest risk for SRO. This finding was in opposition to field studies in claypan watersheds. watersheds.
Figure 6. Particle Adsorbed Figure 6. Particle Adsorbed Runoff RiskRunoff Risk
Day 1 Day 30Day 7 Day 100 Day 180
Particle Adsorbed Runoff Risk at Day 0
Metolachlor
Glyphosate
Atrazine
DKN
Figure 4. Leaching Figure 4. Leaching Risk Risk
Day 1 Day 30Day 7 Day 100 Day 180
Leaching Risk at Day 0
Metolachlor
Glyphosate
Atrazine
DKN
Model LimitationsModel Limitations•Risk at Day 0 was over-estimated for leaching.•Time-dependent functions (degradation, adsorption, hydrolysis, and volatilization) under-estimated dissipation.
•All hydrologic pathways should converge to very low risk by Day 180. For the compounds included in this study, at least 4 half-lives had occurred by Day 180.
•Model sensitivity will be greatest when the number of comparisons are limited.•Temporal comparisons should be limited to the critical transport period of ~60 days after application.
•Herbicide comparisons should be limited to two compounds.
Results and Discussion (con’t.)Results and Discussion (con’t.)
HerbicideLeaching
Surface Runoff
Particle Adsorbed
Runoff
Mean
--------------------------- Score•Days -----------------------------
Atrazine 634† 981 1128 914
DKN 617 964 1022 868
Glyphosate
588 934 1207 910
Metolachlor
587 933 1114 878
Table 1. Time-integrated watershed scores (WTable 1. Time-integrated watershed scores (Wss) for each herbicide and ) for each herbicide and hydrologic pathway. hydrologic pathway.
Figure 5. Surface Figure 5. Surface Runoff RiskRunoff Risk
Day 1 Day 30Day 7 Day 100 Day 180
Surface Runoff Risk at Day 0
Metolachlor
Glyphosate
Atrazine
DKN
Slope (S)
0 10 20 30 40 50
ISR
O W
eig
ht
0
1
2
3
4
5
6
7
8
-Cap at 7 Points-Use uppermost min Ksat in profile-AWC Penalty not applied to graph
100 mmsec-1
10 mmsec-1
1 mmsec-1
0.1 mmsec-1
0.01 mmsec-1
†The lower the Ws, the lower the risk.
Figure 7. Area-weighted risk scores over time for atrazine, DKN, glyphosate, Figure 7. Area-weighted risk scores over time for atrazine, DKN, glyphosate, and metolachlor for each hydrologic pathway. This data was used to compute and metolachlor for each hydrologic pathway. This data was used to compute the watershed scores (Wthe watershed scores (Wss) in Table 1.) in Table 1.
LeachingLeaching SROSRO AROARO
