An Environmental Risk Assessment of Herbicides ...

An Environmental Risk Assessment of Herbicides Used in Channels and Drains in the Goulburn-Murray Irrigation Areas Stage 1: Preliminary Assessment

Ms Nasim Jafari, Dr Rai Kookana, Dr Ray Correll (CSIRO), and Dr Golam Kibria (Goulburn-Murray Water) CSIRO Land and Water science Report 31/08 June 2008 Commercial-in-confidence

CSIRO Land and Water Science Report series ISSN: 1834-6618 G-MW Docs: 2482314

Copyright and Disclaimer © 2008 CSIRO To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO. Important Disclaimer CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it.

Acknowledgements G-MW: Professor Kath Bowmer (Charles Stuart University), and Dr Anne Graesser, Peter Butcher (G-MW) peer reviewed the report. Their comments and suggestions helped to improve the report. A number of G-MW field people supplied data and information. The steering committee members including Mark Finlay, Pat Feehan, Ross Gledhill, Ian Moorhouse, Tim Nietschke, Rod McQueen, and Roger Baker provided valuable inputs during 14 June 2007 discussions in Tatura CSIRO: Ms Natasha Waller, Ms Mary Barnes and Ms Danni Oliver are thanked for their helpful discussions during the project and in improving the structure of the document

For further Contact G-MW: Dr Golam Kibria ([email protected]) CSIRO: DR Rai Kookana ([email protected])

Cover Photograph: Description: Photos of irrigation channels and drain in Goulburn-Murray: Golam Kibria © 2008 CSIRO

Executive Summary Goulburn-Murray Water (G-MW) is the largest rural water supply authority in Australia, supplying water for irrigation, domestic and stock drinking and for raw town supplies. GMW region covers 68,000 square kilometres from the Great Dividing Range north to the River Murray and from Corryong down river to Nyah near Swan Hill. In the irrigation areas along the lower Goulburn and Murray Rivers, G-MW conveys water to and from customers’ properties via earthen channels or drains. These channels and drains can become choked with a variety of aquatic plants. G-MW commonly uses herbicides, mainly glyphosate, amitrole, 2,4-D amine, acrolein, dalapon and imazapyr to control aquatic plants in the irrigation channels, drains and natural carriers so that normal water flow can be maintained. While application of these herbicides is aimed to control targeted plants species, there remains a possibility of potential harm to other non-targeted species if the applications are not properly managed. There has also been concern about herbicide use and its possible impact on the environment, including beneficial use of the waters. In particular, there are concerns that herbicide residues might adversely affect fish and other aquatic biota. In response to the fish deaths in Goulburn River, the Victorian Government commissioned an independent audit of the management of the Goulburn River. The audit recommended that G-MW assess the risks associated with chemical control of aquatic plants in its channels and drains that outfall to the Goulburn River. This report makes a preliminary assessment of some of these risks. The current herbicide risk assessment is based on several scenarios that were jointly identified by G-MW and CSIRO as a part of this collaborative research project. The objective of the assessment was to assess the potential impact of herbicides used by GMW (namely glyphosate, amitrole, dalapon, 2,4-D amine, imazapyr and acrolein) on a range of beneficial uses of water (eg. human health, stock watering, food processing industry, aquaculture, aquatic biota (riverine and wetlands), irrigation of pastures and selected crops). The assessment considers different exposure pathways (eg. spraying on to plants in the channel and subsequent wash-off, spraying onto the soil in a channel or a drain, spraying on channel and drain embankments and then run-off occurring into the channel or drain and spray drift), and also involves use of different receptor organisms, namely algae, Daphnia, fish, rats, drinking water standards and crops (pasture, tomatoes) as surrogates for different types of the beneficial uses listed above. The report has largely drawn data from literature on environmental fate of pesticides and the ecotoxicological concentrations (e.g. NOAELs, EC50) for various receptor organisms. These data together with application rates of herbicides are used to predict likely environmental concentrations under the selected scenario and hazard quotients (HQs) are calculated at the point of treatment. A HQ represents a ratio of predicted environmental concentration (PEC) of a herbicide to threshold ecotoxicological concentration (e.g. NOAELs, EC50) for a receptor organism. The results do not include the dilution and other factors downstream of treated area, which may lead to major reduction in the actual risk to the receiving environment. A tiered (or phased) approach has been adopted in this risk assessment study. In Tier 1 risk assessment, a near-worst case scenario was considered which did not include losses or mitigating factors such as pesticide degradation or plant interception. In many cases the risks were sufficiently low, but the study also identified risks that needed

#2482314 File: 2006/1480/1

ii

further consideration. This required a Tier 2 risk assessment. In Tier 2, a more refined and realistic assessment of the PEC was considered for each herbicide. The estimation of the PEC considered dilution, volatilization, adsorption, biodegradation and effective half-life of the herbicide in each compartment (e.g. water, soil and plant), an estimate of the fraction that would run-off from the bank and the fraction of herbicide applied to a plant that would have washed off from the plant after a channel was refilled with water. In the case of acrolein, hydrodynamic dispersion and diffusion models from literature were adapted to predict PEC. Both Tier 1 and Tier 2 risk assessments considered toxicity values of surrogate or actual receptors organisms such as EC50 (plants, algae or crops), LD50 (rats) and LC50 (Daphnia, fish), and NOAEC (or NOAEL) values. The criteria used in both Tier 1 and Tier 2 risk assessments was obtained by dividing the PEC by NOAEC (or other toxicological value) to derive a HQ. Scenarios with an HQ less than 1.0 were considered not to be at risk and hence did not require further consideration. However, if the HQ from the Tier 1 analysis exceeded 1.0, then a Tier 2 assessment was required. The key results from these assessments are summarized below. 1. Glyphosate did not represent a risk to most receptors considered in this report, even when it was applied at the highest levels of 40 L/ha in G-MW channels. However, the use of glyphosate in drains, could pose a threat to aquatic life, unless the drain water was sufficiently diluted with uncontaminated water (>10 fold). The water from the drains if used for irrigation may have the potential to harm crops. It is noteworthy that these assessments were based on an assumption of high wash-off factor (60% of herbicide residue sprayed on plant is washed off in water). The actual extent of the wash-off may be a lot smaller and needs to be investigated. 2. The application of 2,4-D amine at the highest level of 10L/ha was assessed to pose no acute toxicity risk for mammals, algae and Daphnia, but could harm fish, crops (tomato, soybean) and aquatic ecosystem. 3. Dalapon could potentially harm crops (corn and possibly others) and aquatic plants (duckweed) under conditions assessed in this report but posed no threat to animals. 4. Application of imazapyr could adversely impact algae and duckweed, but posed no toxicological threat to animal species. 5. The application of amitrole in drains could potentially harm fish, irrigated crops and aquatic ecosystems but were assessed to be safe for algae, mammals and Daphnia. 6. Two rates of application of acrolein were assessed for the flow velocity supplied by G-M W (463 m/hr). When acrolein is applied at a rate of 3 mg/L, it was only considered safe to draw the water from the treated channel for irrigation and aquatic life protection at least 8 km and 30 km downstream of the injection point, respectively. In contrast when acrolein was applied at 0.3 mg/L, these distances were reduced to 1 km and 10 km downstream from the injection point, respectively for the two uses. #2482314 File: 2006/1480/1

iii

Discussion and Implications In Tier 2 assessment, herbicide properties such as half-life in water, soil and plant, and wash-off factors were incorporated to estimate the HQ of each receptor. Modeling used in Tier 2 predicted concentrations that were consistent with some monitoring data from the audit, adding credibility to the assessments from this study. The Tier 2 risk assessment showed that application of 2,4-D amine and amitrole (at the current application rate) in G-MW channels and drains could cause harm to some receptors such as fish and certain crops (e.g. soy bean in case of 2,4-D amine). PEC for 2,4-D exceeded the irrigation water quality guideline and may require a 100-fold dilution to meet this guideline. The PEC for amitrole in drains exceeded the irrigation water guideline even when a low wash-off factor of 5% was considered. Irrigation water extracted from a drain that has been treated with amitrole has the potential for causing harm to crops, if used without any dilution. In this case even higher dilution factor (perhaps1000 fold) may be required to meet that guideline depending on the wash-off factor assumed from plants. The degree of dilution should be factored in prior to making decision on the use of drainage water. Such water should not be used for drinking purposes. In contrast, the assessment showed that application of glyphosate is likely to cause minimal effects to most receptors considered in this study. The assessment for acrolein (for the specified velocity of water in channel) found that channels water injected with acrolein would be acceptable for the purpose of irrigation, aquatic life protection and others usage, if the water is drawn/used beyond the distances specified in this report form the herbicide injection point. When acrolein is used to control weeds in channels, it should be used at the lower application rate of 0.3 mg/L to minimize the environmental impact, and that the water should be permitted to run in a channel to reach a “safe distance” as specified in this report for various receptors (e.g. drinking water, aquatic life, crop irrigation). These distances should be calculated for the specific case under consideration by taking the velocity of water in the channel into account, which is one of most important parameters affecting “safe distance”. The current assessment does not take into account the degree of dilution that may occur downstream of the treated area. Therefore, the data and information need to put into the context of actual use scenario. These considerations include size of channel or drain, depth of water, dilution factor and time before the water may reach to the ecosystem being protected. The current herbicide risk assessment was based on assumed wash-off values for the herbicide residue from the plant surface into water. While the assumed values covered the feasible range, it was found that the risk assessment was very sensitive to these values and typically the highest value was assumed. It is therefore recommended that estimates of the wash-off fractions should be obtained initially using laboratory experiments and then verified in field situations. Secondly, there are also little data available on adsorption of pesticides by the soil at the bottom of the channels and drains. Such data should be collected from samples taken from channels and drains in the G-MW region. Both the wash-off data and the soil absorption data should be incorporated into the risk assessment, prior to making specific decisions. A summary of the results is given in Table 1 where maximum values have been used for wash-off, and acrolein concentrations were taken using the application rate of 3 mg/L but with the results assessed 5 km down stream. #2482314 File: 2006/1480/1

iv

Table 1. Overall Summary of Hazard Quotients, HQs (Red HQ > 1, amber HQ 0.1 – 1.0 and green HQ < 0.1) using typical values. For the conditions, application rates and surrogate receptors for a particular use please refer to main body of the report. Location

Receptor

Glyphosate

2,4-D

G-MW Channels

Humans Drinking

0.61

Stock Drinking

Amitrole

Imazapyr

Dalapon

Acrolein*

17

33

4.44

1.81

0.00

0.4

0

0.02

6.4

Irrigation

4.3

40

760

222

Crops

0.20

7.9

0.17

2.54

0.39

0.05

73

0

0.89

51

0.05

73

0.00

0.01

51

0.89

580

Aquaculture Fish Riverine ecosystems and wetlands Macroinvertebrates

G-MW Drains

0.43

0.2

0.00

0.89

Fish

0.05

73

0.00

0.01

Algae

0.09

0.045

1.55

0.06

Birds

0.00

0.00

0

Water for human and stock consumption

2.2

11600

570

Pastures

0.70

Outfall to rivers & wetlands

1.3

2.85

Irrigation

15

285

1.3

2.85

Aquatic ecosystems

* Acrolein assessment is based on the application rate of 2.58 mg/L in water (flowing at velocity of 463 m/hr), with the results assessed 5 km down stream of application point.

#2482314 File: 2006/1480/1

v

Recommendations 1. Before making any changes to management practices, the findings in this report should be carefully reviewed for each particular scenario under consideration. Such an analysis should incorporate dilution factors and other risk mitigating or moderating factors such as channel/drain dimensions (especially depth of water), buffer between treated area and receiving environments, flow velocity and the withholding periods. 2. Amitrole was found to pose the highest risk, so we agree with G-MW recent decision to discontinuing the use of amitrole. 3. Accurate data should be obtained for the wash-off fractions under the conditions relevant to this study. Inclusion of such data is expected to provide more realistic assessments and may lower the estimated risk. Data should also be obtained on the binding ability of pesticides to the channel and drain floor. These data would complement the wash-off fraction data mentioned above. 4. The ongoing monitoring of herbicide residues as recommended in the Goulburn River Audit (2005) should continue and include a time series for the concentrations of pesticides in channels and drains following spraying and refilling the irrigation channel. Strategic monitoring following specific herbicide treatment is needed to further validate the current desk top study. 5. The potential use of imazapyr should continue to be explored in conjunction with monitoring of its rate of degradation in a refilled channel. That study should take into account photo-degradation of the imazapyr. 6. Data should be obtained on the effects of low concentrations of the key herbicides on selected crops (including corn) to enable more realistic levels of the permitted concentration of herbicides in irrigation water. 7. The data presented in this report (on acceptable concentrations and environmental fate of herbicides) could be utilised to develop an interactive EXCEL based risk assessment tool that can be used to calculate risk for the specific scenario being assessed. Such a model could include parameters such as water depth and channel width.

#2482314 File: 2006/1480/1

vi

Table of Contents Executive Summary

ii

Recommendations

vi

Table of Contents

vii

List of Figures

xiii

Acronyms and glossary of terms

xiv

Acronyms and glossary of terms

xiv

1.

1

2.

Introduction 1.1.

Aquatic Plant Control

2

1.2.

Scope of this Study

2

Herbicide use in Channels and Drains 2.1.

Herbicide Use Scenarios

5

2.2.

Channels in the Murray Valley and Shepparton Areas

5

2.2.1.

Use of glyphosate in the Murray Valley and Shepparton Areas

6

2.2.2.

Use of 2,4-D amine in the Murray Valley and Shepparton Areas

7

2.2.3.

Use of imazapyr in the Murray Valley Area

7

2.2.4.

Use of acrolein in Murray Valley

7

2.3.

Channels in the Central Goulburn and Rochester-Campaspe Areas

2.3.1. 2.4.

2.5.

7

Use of glyphosate in the Central Goulburn and Rochester-Campaspe Areas .......8

Channels in the Pyramid Boort Area

2.4.1.

Use of glyphosate in the Pyramid Boort Area

Channels in the Torrumbarry Area

9 9 9

2.5.1.

Use of glyphosate in the Torrumbarry Area

9

2.5.2.

Use of dalapon in the Torrumbarry Area

9

2.5.3.

Use of acrolein in the Torrumbarry Area

10

2.6.

3.

5

Drains in the G-MW Irrigation Region

10

2.6.1.

Use of glyphosate in drains

10

2.6.2.

Use of amitrole in drains

10

Environmental fate and toxicological properties of the herbicides 3.1.

Glyphosate

11 11

3.1.1.

Toxicity of glyphosate to different receptors

11

3.1.2.

Degradation in water

11

3.1.3.

Degradation in soil

12

3.1.4.

Run-off of glyphosate

12

3.1.5.

Degradation of glyphosate in plants

12

3.2.

Amitrole

#2482314 File: 2006/1480/1

13

vii

3.2.1.

Toxicity of amitrole to different receptors

13

3.2.2.

Degradation of amitrole in water

13

3.2.3.

Degradation of amitrole in soil

14

3.2.4.

Run-off of amitrole

14

3.2.5.

Degradation of amitrole in plants

14

3.3.

2,4-D amine

3.3.1.

Toxicity of 2,4-D amine to different receptors

14

3.3.2.

Degradation of 2,4-D in water

15

3.3.3.

Adsorption and degradation of 2,4-D in soil

15

3.3.4.

Run-off of 2,4-D

16

3.3.5.

Degradation of 2,4-D in plants

16

3.4.

Dalapon (2,2 DPA)

16

3.4.1.

Toxicity of dalapon to different receptors

16

3.4.2.

Adsorption and breakdown in soil

16

3.4.3.

Run-off

16

3.4.4.

Adsorption and breakdown in water

17

3.4.5.

Absorption and breakdown in plants

18

3.5.

Imazapyr

3.5.1.

Toxicity of imazapyr to different receptors

18 18

* 15% mortality

19

3.5.2.

Adsorption and breakdown in soil

19

3.5.3.

Run-off

19

3.5.4.

Adsorption and breakdown in water

20

3.5.5.

Absorption and breakdown in plants

20

3.6.

Acrolein

20

3.6.1.

Toxicity of acrolein to different receptors

20

3.6.2.

Adsorption and break down of acrolein in water

21

3.6.3.

Degradation of acrolein in plants

21

3.7. 4.

14

Summary of pesticide properties

Methodology used for Risk Assessment

22 23

4.1.

Terms used in estimation

23

4.2.

Tier 1 Assessment of Risk

24

4.2.1. 4.3.

Tier 1 Principles

Tier 1 Estimation of PEC

24 25

4.3.1.

Calculation process of PEC for herbicide applied directly to a channel

25

4.3.2.

Estimation of PEC for Acrolein

27

4.4.

Tier 2 Estimation of PEC

#2482314 File: 2006/1480/1

28

viii

4.4.1.

Processes considered in Tier 2

30

4.4.2.

Interception of herbicides by plants

31

4.4.3.

Wash-off of herbicides from plants

31

4.4.4.

Herbicide sprayed onto soil

32

4.4.5.

Run-off from the bank

32

4.4.6.

Mass of soil

33

4.4.7.

Total pesticide burden

33

4.4.8.

Sorption of pesticides by soil

33

4.5.

Tier 2 calculations for acrolein

4.5.1.

5.

35

4.6.

Estimation of Drinking Water Equivalent Level (DWEL)

36

4.7.

Estimation of Hazard Quotients

36

Results 5.1.

5.2.

38

Estimation of Tier 1 concentrations in channels (excluding acrolein)

5.1.1.

Tier 1 estimation of concentrations in drains

Estimation of Tier 1 hazard quotients

38 38 39

5.2.1.

Tier 1 hazard quotients for channels (excluding acrolein)

39

5.2.2.

Tier 1 hazard quotients for channels for acrolein

43

5.2.3.

Tier 1 hazard quotients for drains

44

5.3.

Estimation of Tier 2 concentrations

45

5.3.1.

Tier 2 Estimation of concentrations in channels (excluding acrolein)

45

5.3.2.

Tier 2 estimates of acrolein concentration

48

5.3.3.

Tier 2 estimates of concentrations of herbicides in drains

50

5.4.

6.

Dispersion

35

Estimation of Tier 2 hazard quotients

50

5.4.1.

Tier 2 hazard quotients for channels (excluding acrolein)

50

5.4.2.

Tier 2 hazard quotients for acrolein in channels

55

5.4.3.

Tier 2 hazard quotients for drains

58

Discussion

60

6.1.

Comparison of predictions with monitoring data

60

6.2.

Comparison of herbicide effects

60

6.2.1.

Glyphosate

60

6.2.2.

2,4-D amine

61

6.2.3.

Amitrole

61

6.2.4.

Dalapon

61

6.2.5.

Imazapyr

62

6.2.6.

Acrolein

62

6.2.7.

Overview

62

#2482314 File: 2006/1480/1

ix

6.3.

Extrapolation to other receptors

62

6.4.

Other risk not considered in this report

63

6.4.1.

Acute versus chronic risks

64

6.4.2.

Daughter products

64

6.5. 7.

Closing remarks

References

65 66

Appendices

72

#2482314

x

File: 2006/1480/1

List of Tables Table 1. Overall Summary of Hazard Quotients, HQs

v

Table 2. Areas of districts within the GM-W Irrigation Region.

2

Table 3. Summary of various scenarios assessed during the risk assessment

3

Table 4 Summary of herbicide use.

5

Table 5. Example of detail use of acrolein application

7

Table 6. Toxicity of glyphosate to different receptors

11

Table 7 Toxicity of amitrole to different receptors

13

Table 8. Toxicity of 2,4-D amine to different receptors

14

Table 9. Ecotox values for different receptors for dalapon

17

Table 10. NOAEC of different receptors for Imazapyr

18

Table 11. Toxicity of Acrolein to different receptors

20

Table 12 Properties of herbicides used by G-MW and considered in this report

22

Table 13 Mathematical symbols used in this section

23

Table 14 Calculation of PEC.

27

Table 15 Conversion of volumetric to gravimetric concentration for acrolein

28

Table 16. PEC estimations in Tier 1 for channel for glyphosate, 2,4-D, imazaypyr and dalapon. 38 Table 17. PEC estimations in Tier 1 for drains for glyphosate and amitole

38

Table 18. Estimation of Tier 1 Hazard Quotient for glyphosate applied to channels.

39

Table 19. Estimation of Tier 1 Hazard Quotient for 2,4-D applied to channels.

40

Table 20. Estimation of Tier 1 Hazard Quotient for dalapon applied to channels.

41

Table 21. Estimation of Tier 1 Hazard Quotient for imazapyr applied to channels.

42

Table 22. Tier 1 assessment of the risk posed by acrolein applied at two different methods. ...... 43 Table 23. Tier 1 assessment of the risk posed by glyphosate to drains.

44

Table 24. Tier 1 assessment of the risk posed by amitrole applied to drains.

45

Table 25 Estimation of concentration of glyphosate in channel water.

46

Table 26. Estimation of concentration of imazapyr in channel water.

47

Table 27. Predicted environmental concentrations in channels of 4 herbicides.

48

Table 28. Predicted environmental concentrations in drains of 4 herbicides.

50

Table 29. Tier 2 hazard quotients for glyphosate applied to channels.

51

Table 30. Tier 2 hazard quotients for 2,4-D applied to channels.

52

Table 31. Tier 2 hazard quotients for dalapon applied to channels.

53

Table 32. Tier 2 hazard quotients for imazapyr applied to channels.

54

Table 33. Hazard Quotient for 9 receptors at different distances from source of application of 3.0 mg/L of acrolein. 56 Table 34. Hazard Quotient for 9 receptors at different distances from source of application of 0.3

#2482314 File: 2006/1480/1

xi

mg/L of acrolein.

57

Table 35. Tier 2 hazard quotients for glyphosate applied to drains.

58

Table 36. Tier 2 hazard quotients for amitrole applied to drains.

59

Table 37 Range of monitored data and predicted values for glyphosate, 2,4-D and amitrole ...... 60 Table 38. Comparison of tolerance values across receptors

63

Table 39. Comparison of near worst risks posed by of herbicides to key receptors when applied at maximum rates 63 Table 40. Toxicity of components of Roundup® to rainbow trout

64

#2482314

xii

File: 2006/1480/1

List of Figures Figure 1. Map showing location of G-MW Region together with the irrigation areas under its control. ..............................................................................................................................1 Figure 2. Typical cross-section of an irrigation channel ...................................................................6 Figure 3. Patches of cumbungi in an irrigation channel. ..................................................................6 Figure 4 Diagrammatic steps involved in a Tier 1 assessment .................................................... 25 Figure 5. Diagrammatic cross-section of channel ......................................................................... 26 Figure 6. Comparisons of when bank is sprayed or not sprayed. ................................................. 26 Figure 7. Diagrammatic steps involved in a Tier 2 risk assessments ........................................... 29 Figure 8. Method for assessing the concentration of pesticide in water following spraying with herbicide. ....................................................................................................................... 29 Figure 9. Water status in channels and drains at the time of herbicide application. ..................... 30 Figure 10. Estimated concentrations of acrolein at different times and distances from the point of injection following an injection at 2.58 mg/L for a channel 12 m wide and 1.5 m deep and a water velocity of 463 m/hour ............................................................................... 49 Figure 11. Estimated concentrations of acrolein at different times and distances from the point of injection following an injection at 0.258 mg/L for a channel 12 m wide and 1.5 m deep and a water velocity of 463 m/hour ............................................................................... 49

#2482314 File: 2006/1480/1

xiii

Acronyms and glossary of terms Acute toxicity: Adverse effects occurring within a short time of administration of a single dose of a chemical, or immediately following short or continuous exposure, or multiple doses over 24 hours or less ARMCANZ: Agriculture and Resource Management Council of Australia and New Zealand Bioaccumulation: The absorption, via breathing, eating, drinking or active uptake, and concentration of a substance in plants or animals Bioavailability: The extent to which a substance is capable of entering into biological metabolism Biodegradation: It is the process by which organic substances are broken down by other living organisms. The term is often used in relation to ecology, waste management, environmental remediation (bioremediation) and to plastic materials, due to their long life span. Organic material can be degraded aerobically, with oxygen, or anaerobically, without oxygen. A term related to biodegradation is biomineralisation, in which organic matter is converted into minerals.Biodegradable matter is generally organic material such as plant and animal matter and other substances originating from living organisms Boom sprayer: The most commonly used equipment for applying pesticides is the tractorpowered boom sprayer fitted with conventional hydraulic nozzles. Conventional boom sprayer consists of spray tank, pump, and a boom to which is fixed the spray line, droppers and nozzles. Channels: Open channel or flume designed to convey water from upstream source to farms. Supply channels can be categorised as Main channels whose primary purpose is to convey bulk water from headworks storage or river diversion point into the distribution system; or Distribution channels whose primary purpose is to deliver water from main channels to individual farms Chronic: Long term, low level exposure to a toxic chemical Concentration: The amount of active ingredient or pesticide equivalent in a quantity of diluent Contaminant: Any biological, chemical, physical and radiological substance or matter that has an adverse response (effect) on air, water, soil or living things CSIRO: Commonwealth Scientific and Industrial Research organisation Degradation: A chemical alteration to a pesticide. Chemical or biological breakdown of a complex compound into simpler compounds Domestic & stock water (D&S): This is a small water entitlement, for supplying households, watering of cattle and other stock, water of animals kept as pets. D&S water is untreated and are not to be used for human consumption or drinking EC50 or Median effective concentration: The concentration of material in water that is estimated to be effective in producing some lethal response in 50% of the test organisms Ecosystem: Community of organisms interacting with each other and the chemical and physical factors making up their environment End-point: In toxicity testing and evaluation it is the adverse biological response in question that is measured. End points vary with the level of biological organization being examined but include changes in biochemical markers or enzyme activities, mortality, or survival, growth, reproduction, primary production, and changes in structure (and abundance) and function in a community. End points are used in toxicity tests as a criteria for effects Environmental fate: The destiny of a chemical after release to the environment; involves #2482314 File: 2006/1480/1

xiv

considerations such as transport through air, soil and water, bio-concentration, degradation, etc EPA: Environment Protection Authority Exposure: The amount of physical or chemical agent that reaches a target or receptors Foliage: Leaf, the main organ of photosynthesis and transpiration in higher plants G-MW: Goulburn Murray Rural Water Authority Goulburn River: The Goulburn River is a major inland river in Victoria, Australia. The headwaters of the Goulburn River rise in the western end of the Victoria Alps, near Mount Buller. The Eildon Dam creates Lake Eildon, a major storage of water for irrigation. From Lake Eildon, most of the irrigation water goes to Goulburn Weir and Waranga Basin. North of Eildon the Goulburn River enters the northern plains of Victoria and eventually flows into the Murray River near Echuca. This area is a very productive irrigated agricultural area. The Goulburn River was named after Henry Goulburn. There is also a Goulburn River in New South Wales. Guideline trigger values: These are the concentrations (or loads) of the key performance indicators measured for the ecosystem, below which there exists a low risk that adverse biological (ecological) effects will occur. They indicate a risk of impact if exceeded and should ‘trigger’ some action, either further ecosystems specific investigations or implementation of management /remedial actions Half-life (T ½): The time required for half of the residue to lose its analytical identity whether through dissipation, decomposition, metabolic alteration or other factors Hand guns: This type of applicator commonly used for tree crops or spot spraying of herbicides consists of a hand-held gun or wand fed from a large container on a vehicle though a long hose. The pesticide is pumped from the drum by a motor driven pump or tractor pto. The gun may consists of 1-3 nozzles. Hazard: (1) Likelihood that exposure to a chemical will cause an injury or adverse effect under the conditions of its production, use or disposal; (2) The potential or capacity of a known or potential environmental contaminant to cause adverse ecological effects Hazard Quotient (HQ): The ratio of estimated site-specific exposure to a single chemical from a site over a specified period to the estimated daily exposure level, at which no adverse health effects are likely to occur. Or Hazard quotient means the value which quantifies non-carcinogenic hazard for a single chemical for an individual receptor over a specified exposure period. The hazard quotient is equal to the ratio of an intake of a chemical to the chemical's reference dose. Hazard quotient shall be based on similar-acting non-carcinogens, i.e., systemic toxicants that act on the same organ or organ system Hydrolysis: Decomposition of a chemical compound by reaction with water, such as the dissociation of a dissolved salt or the catalytic conversion of starch to glucose Irrigation: Irrigation is the artificial application of water to land for the purpose of agricultural production. Effective irrigation will influence the entire growth process from seedbed preparation, germination, root growth, nutrient utilisation, plant growth and regrowth, yield and quality Koc or adsorption coefficient: A measure of a materials tendency to adsorb to soil particles. High Koc indicate a tendency for the material to be adsorbed by soil particles rather than remain dissolved in the soil solution. In general strongly adsorbed molecules will not leach. Koc values of 2.5 are hydrophobic #2482314 File: 2006/1480/1

xv

LC50 or median lethal concentration: The concentration of material in water that is estimated to be lethal to 50% of the test organisms. The LC50 is usually expressed as a time-independent value, e.g. 24 hour or 96- hour LC50, the concentration estimated to be lethal to 50% of the test organisms after 24 or 96 hours of exposure LD50 or median lethal dose: The concentration of material in water that is estimated to be lethal to 50% of the test organisms. Appropriate to use with test animals such as rats, mice and dogs. It is rarely applicable to aquatic organisms because it indicates the quantity of a material introduced directly into the body by injection or ingestion rather than the concentration of the material in water in which aquatic organisms are exposed during toxicity tests LOEL: Lowest Observed Effect Level; the lowest dose in an experiment which produced an observable effect Mammals: The class of organisms that have backbones (vertebrates); includes all animals that have hair and suckle their young MCL: Maximum contaminant level Murray River: The Murray River, or River Murray, is Australia's second-longest river in its own right (the longest being its tributary the Darling). At 2,575 kilometres (1,600 miles) in length, the Murray rises in the Australian Alps, draining the western side of Australia's highest mountains and, for most of its length, meanders across Australia's inland plains, forming the border between New South Wales and Victoria as it flows to the northwest, before turning south for its final 500 kilometres or so into South Australia. The waters of the Murray flow through several lakes that fluctuate in salinity (and were often fresh until recent decades) including Lake Alexandrina and The Coorong before emptying through the Murray mouth into the Indian Ocean (Southern Ocean according to Australian maps) near Goolwa. Despite discharging considerable volumes of water at times, particularly before the advent of large scale river regulation, the Murray mouth has always been comparatively small and shallow. NOEL: No observed adverse effect level-the highest tested concentration at which no adverse effect was observed NOEC: No observable effect concentration Octanol-water partition coefficient (Kow): The ratio of a chemicals solubility in n-octanol (C8H17OH) to its solubility in water. Symbol Kow. The ratio indicates the chemicals propensity for bio-concentration by aquatic organisms. It is an important parameter and is used often in the assessment of environmental fate and transport for organic chemicals Outfall: Regulating structure located at the downstream end, or intermediate points, of a supply channel to allow safe discharge of surplus flows arising in the system due to the effects of rainfall inflow, planned channel shutdown or operational error. An outfall can also be used to drain water from the channel at the end of the irrigation season. Water released through the outfall is usually discharged to a drainage channel, natural waterway or Regulating Storage Partition coefficient: A ratio of the equilibrium concentration of the chemical between a nonpolar and polar solvent Pesticides: Any chemical compound used to kill pests that destroy agriculture production or are in someway harmful to humans. Pesticides include herbicides (eg. 2, 4-D) which kill unwanted plants or weeds; insecticides (eg. endosulfan) which kill insect pests; fungicides (eg. copper hydroxide) which kill fungi Photolysis: Chemical decomposition induced by light or other radiant energy Predicted Environmental Concentration (PEC): The Predicted Environmental Concentration is an indication of the expected concentration of a material in the environment, taking into account the amount initially present (or added to) the environment, its distribution, and the probable methods and rates of environmental degradation and removal, either forced or natural.

#2482314 File: 2006/1480/1

xvi

PTO: Power take off, typically from the engine of a tractor Rate: The amount of active ingredient or acid equivalent applied per unit area or other treatment Receptor: Receptor means environmental resources, including but not limited to, plant and animal species, humans, sensitive environments and habitats, water supply wells, and locations that have the potential to be, or have actually been, exposed to contamination. In exposure assessment, an organisms that receives, may receive, or has received environmental exposure to a chemical Residue: That quantity of pesticide, its degradation products, and/or its metabolites remaining on or in the soil, plant parts, animal tissues, whole organisms, and surfaces Risk: A statistical concept defined as the expected likelihood or probability of undesirable effects resulting from a specified exposure to known or potential environmental concentrations of a material. A material is considered safe if the risks associated with its exposure are judged to be acceptable Risk assessment: A qualitative or quantitative evaluation of the environmental and/ or health risk resulting from exposure to a chemical or physical agent (pollutant). Risk assessment or "sitespecific risk assessment" means a site-specific characterization of the current or potential threats that may be posed to human health and the environment by contamination migrating to or in groundwater or surface water, discharging to the air, leaching through or remaining in soil, bioaccumulating in the food chain, or other complete and significant exposure pathways identified in the Site Conceptual Exposure Model (SCEM). Key components of a risk assessment are the identification of hazard (i.e., identifying site-related chemicals and their concentrations in the exposure media), exposure assessment (identifying complete and significant exposure pathways and quantifying intake), toxicity assessment (identifying the toxic effects and dose-response [toxicity value]), risk characterization, and discussion of uncertainties. For the purposes of these regulations, a Tier 3 Risk Assessment is considered a "site-specific risk assessment." River: A river is a large natural waterway. The source of a river may be a lake, a spring, or a collection of small streams, known as headwaters. From their source, all rivers flow downhill, typically terminating in the ocean. The mouth, or lower end, of a river is known as its base level. Run-off: That portion of precipitation which is not absorbed into the soil, but flows into surface streams Soil mobility: Movement of a compound through soil from the treated area by leaching, volatilization, adsorption and desorption or dispersal by water Sub-lethal: Having an effect less severe than death Surface water: Water in open bodies such as streams, rivers, ponds, lakes and oceans Toxicants: substances that is harmful to living organisms Target: The target species is the organism which the pesticide is intended to control Tiered approach for risk assessment: A tiered assessment is a risk assessment that is set up in a number of sequential steps of increased complexity and effort and there are specifies decision criteria for each step. It involves evaluating whether or not the next step of assessment should be undertaken based on these criteria (Figure source: USEPA) #2482314 File: 2006/1480/1

xvii

Toxicant: An agent or material capable of producing an adverse response (effect) in a biological system, seriously injuring structure and/or function or producing death Toxicity: The inherent potential or capacity of an agent or material to cause adverse effects in a living organism when the organism is exposed to it Toxicity test: A measure of the degree of response of an organism exposed to a particular concentration of a chemical or a particular level of some other environmental variable Uptake: A process by which materials are absorbed and incorporated into a living organism Wash-off: The removal of herbicides applied on a plant that is washed off by irrigation water

#2482314 File: 2006/1480/1

xviii

1.

Introduction

Goulburn-Murray Water (G-MW) is the largest rural water supply authority in Australia, supplying water for irrigation, domestic and stock drinking and for raw town supplies. G-MW’s region covers 68,000 square kilometres from the Great Dividing Range north to the River Murray and from Corryong down river to Nyah near Swan Hill (see Figure 1, below). G-MW’s retail business comprises 2 main types of service. The Diversions business involves servicing customers who take water from streams, lakes, wetlands and aquifers throughout the region using their own infrastructure (e.g. pumps and bores). The Gravity business involves conveying water to customer properties in irrigation channels and conveying water away from those properties in drains. This report deals exclusively with the Gravity business, which is confined to 6 Irrigation Areas (see coloured areas in Figure 1, below) as detailed in Table 2.

Figure 1. Map showing location of G-MW Region together with the irrigation areas under its control. Map is available from http://www.g-mwater.com.au/browse.asp?ContainerID=area_map

#2482314 File: 2006/1480/1

1

Table 2. Areas of districts within the GM-W Irrigation Region, together with the lengths of channels and drains Irrigation Area

Area (ha)

Irrigated area (ha)

Length of Channels (km)

Length of Drains (km)

Pyramid Boort

166,215

126,400

1302

111

Torrumbarry

167,000

150,000

1385

708

Rochester-Campaspe

117,050

66,710

599

507

Central Goulburn

173,053

113,106

1353

892

Shepparton

81,750

51,000

576

444

Murray Valley

128,372

88,969

1041

484

Total

596,185

596,185

6256

3146

G-MW owns and operates 6,256 km of irrigation channels and 3,146 km of drains in its Irrigation Areas. Channels vary in capacity from 10 to 3,500 Megalitres (ML) per day within an Irrigation Area. The drainage network carries excess irrigation water from farms and conveys it to the river system or to wetlands, with irrigation reuse from G-MW drains becoming increasingly important.

1.1. Aquatic Plant Control A key challenge for G-MW is to control the growth of aquatic plants in its channels and drains to maintain design flows without causing significant downstream impacts. Aquatic plant control is generally achieved by manipulating water depth and the use of five herbicides, namely acrolein, glyphosate, amitrole, 2,2 DPA and 2,4-D amine. Application of these herbicides on target species may affect non-target receptors as well. G-MW is aware of the potential for collateral damage and has therefore commissioned this report to assess the risk of harm to a range of non-target species (Kookana et al., 2003). There has also been increasing public concern over herbicides use and its possible impact on beneficial water uses. In particular, there are concerns that herbicide residues might cause fish deaths, such as those that occurred in the Goulburn River downstream of Goulburn Weir (see Figure 1, above) in January 2004. In response to the Goulburn River fish deaths, the Victorian Government commissioned an independent audit of the management of the Goulburn River (EPA 2005). The audit recommended that G-MW assess the risks associated with chemical weed control in its channels and drains that outfall to the Goulburn River.

1.2. Scope of this Study The current study was to assess seven scenarios throughout G-MW’s Irrigation Areas (see Table 3 below) that were identified jointly by G-MW and CSIRO, using intellectual property and tools available from CSIRO, through development of new methods for risk assessment, and utilizing G-MW’s raw data and information. This assessment therefore exceeds the Goulburn River Audit (EPA 2005) recommendations in as much as it assesses herbicide residue risks for rivers other than the Goulburn and also assesses risks associated with direct use of channel and #2482314 File: 2006/1480/1

2

drainage water. The main receptors that were considered in that brief were the water flea (Daphnia sp.) representing aquatic invertebrates, rainbow trout (Oncorhynchus mykiss) representing fish, algae representing the base of the foodchain and a range of crops (including tomatoes and pastures) as well as human and stock consumption. These receptors were chosen in consultation with G-MW to represent the various ways the water is used in the region. This study is based on standard risk assessment principles and employs a range of scenarios based on data available for herbicides used in the six G-MW Irrigation Areas. A tiered or phased approach has been used in this study. In the first phase (termed Tier 1 in the risk assessment literature) a near worst case scenario is considered, often with unrealistic conditions to remove those candidates that are not worth further assessment. In this phase, it is assumed that no loss of the herbicides occurs after application either from degradation or absorption. In the second phase (or Tier 2) a more refined assessment of the expected environmental concentrations is used in the risk assessment. The first and second tier assessments both assess the exposure of each receptor organisms likely to be exposed to the hazard (i.e. herbicides). The first tier considers a first approximation where the near worst case is considered at each step. More realistic approximations are used in the Tier 2 assessments. The most sensitive receptors were chosen, in line with the worst case approach taken with the Tier 1 risk assessments. For examples, a previous study had established that native fish were less sensitive to many of the herbicides used by GMW than the exotic species commonly used in ecotoxicological tests (Raymond et al. 2006). Rainbow trout were therefore chosen as representative receptors for aquatic vertebrates, rather than the native fish commonly found in rivers and wetlands in GMW’s region. The validity of the results is dependent on the quality of input data and the assumptions made in the assessments (including modelling assumptions and working approximations). These assumptions and approximations have been discussed in this report where appropriate. In many cases the accuracy of the assessments was limited by the availability of suitable data which needs to be considered before making recommendations both for management practice and for future research. In compiling this report the authors encountered several new terms and common plant names. For completeness there is a full description of terms, the plants etc. considered in this report and an extensive glossary and several appendices have been included. These appendices are useful in their own right and are relevant here, but they have not been restricted to those referenced in this report. Table 3. Summary of various scenarios assessed during the risk assessment Scenario

Source

Transport Pathway

Receptor

1: Direct human consumption and contact (all covered via mammalian toxicity)

G-MW Channels

Direct in water On the channel bed Spraying on channel embankmentinput through run-off Noxious weed sprays on adjacent lands, weirs, access roads

Humans Drinking water Food processing (equipment washing, produce washing) Domestic supply (washing/cooking/bathing) Recreation (swimming, boating, fishing)

#2482314 File: 2006/1480/1

3

2; Direct stock consumption (covered via mammalian toxicityas scenario 1)

G-MW Channels

3: Irrigation of pastures, crops and trees.

G-MW Channels

4: Aquaculture in channel water

G-MW Channels

5; Aquatic ecosystems partially supplied by channel outfalls (covered via mammalian toxicity)

G-MW Channels

6; Aquaculture, stock consumption and irrigation with water taken directly from drains

G-MW Drains

7: All of the above for drainage water that is outfalled to streams and wetlands (and therefore considerably diluted)

G-MW Drains

#2482314 File: 2006/1480/1

Direct in water On the channel bed Spraying on channel embankmentinput through run-off Noxious weed sprays on adjacent lands, weirs, access roads Through food contaminated with herbicides eg. Pastures Direct in water On the channel bed Spraying on channel embankmentinput through run-off Noxious weed sprays on adjacent lands, weirs, access roads Direct in water On the channel bed Spraying on channel embankmentinput through run-off Noxious weed sprays on adjacent lands, weirs, access roads Direct in water On the channel bed Spraying on channel embankmentinput through run-off Noxious weed sprays on adjacent lands, dams, access roads Channels outfall Direct in water On the drains bed Spraying on drains embankment-input through run-off Noxious weed sprays on adjacent lands, access roads Drift Drainage water discharge into rivers and wetlands Direct in water On the drains bed Spraying on drains embankment-input through run-off Noxious weed sprays on adjacent lands, access roads Drift Drainage water discharged into natural waterways

Stock Drinking water Washing of milking equipment etc (all covered via mammalian toxicity)

Irrigation Pasture (1) Tomatoes (1) Crop (wheat) (1)

Aquaculture Fish

Riverine ecosystems and wetlands Macro-invertebrates (1) Fish (1) Algae (1) Other aquatic plants (1) Birds (1) Humans and stocks Humans (1) Stock Pastures (1) Tomatoes (1) Crop (wheat) (1) Aquaculture Fish

Natural waterways Pasture (1) Aquatic ecosystems (5) Humans (1) Stock Pastures (1) Aquaculture Food Processing (equipment washing, produce washing) Domestic supply (washing/cooking/bathing) Recreation (swimming, boating, fishing)

4

2. Herbicide use in Channels and Drains 2.1. Herbicide Use Scenarios Table 4 below, summarizes how herbicides are used in G-MW channels and drains. This summary includes imazapyr, which is being trialed as an alternative control for arrowhead in channels, as well as the five herbicides currently used by G-MW. Table 4 Summary of herbicide use. (Target weeds are described colloquially. Botanical names are given in Appendix B) Herbicide

Trade Name

Target Species

Application Conditions

Application Rates Assessed

Glyphosate

Roundup® Bioactive

Cumbungi, water couch, arrowhead

Emergent weeds in channels and drains

3.2-14.4 kg a.i./ha

2,4-D amine

Amicide LO500A®

Arrowhead, milfoil

Emergent weeds in channels

Up to 6.25 kg a.i./ha

Amitrole

Amitrole T or Amitrole TL

Water couch, barnyard grass, umbrella sedge, broadleaf weeds

Emergent weeds in drains

Up to 2.75 kg a.i./ha

2,2 DPA

Dalapon

Cumbungi

Emergent weeds in channels and drains

7.1 kg a.i./ha

Imazapyr

Arsenal Express

Arrowhead

Used experimentally in channels

0.75 kg a.i./ha

Acrolein

Magnacide H®

Ribbon Weed, pondweed, Elodea

Submerged weeds in channels

0.3 – 3 ppm (mg/L)

A summary of the properties of these six herbicides that are used by G-MW is provided in Section 3. The weed profile of the channels varies among the Irrigation Areas. This in turn implies a different herbicide profile occurs in each Irrigation Area. These differences are briefly discussed in this section.

2.2. Channels in the Murray Valley and Shepparton Areas Murray-Valley (128,372 ha) with 88,969 ha irrigated area includes 1041 km of channels and 484 km of drains. The main land uses in this area are cropping and grazing (45%) and horticulture (stone fruit) (8%). The Shepparton Area occupies 81,750 ha of which 51,000 is irrigated. This area includes 576 km channel and 444 km drain in total. Together there are 1617 km of channels in these two areas, with the channel sizes mainly varying from 3 to 5 m in width and 0.3 to 0.5 m in depth. Some of the trunk channels that are sprayed with 2,4-D amine to control arrowhead, are much larger. A #2482314 File: 2006/1480/1

5

typical cross-section of a channel is shown in Figure 2. As shown in Figure 2, the surface area of the banks may exceed that of the water in the channel. The ratio of the area of the banks to the area of the water is an important factor used in the risk assessments, as sprays applied to the banks have the potential to be washed into the channels. This risk is explored later in this report.

Figure 2. Typical cross-section of an irrigation channel

2.2.1. Use of glyphosate in the Murray Valley and Shepparton Areas Arrowhead Glyphosate is used to control arrowhead when it covers much or the entire channel and restricts the flow of water. Spraying to control arrowhead is usually performed following a decrease in the water level in the channel and under these conditions there is no water movement. The channel bed is then sprayed with glyphosate at rates of 20 - 40 L/ha using a boom spray.

Cumbungi Cumbungi (bull rush) often occurs in small patches of channel and may occupy only 0.1% of channel system. These patches are spot sprayed at a rate of 9 L/ha by hand gun at that early stage to prevent them becoming too established.

Figure 3. Patches of cumbungi in an irrigation channel, where often the cover is as low as 0.1% of area of the channel

#2482314 File: 2006/1480/1

6

2.2.2. Use of 2,4-D amine in the Murray Valley and Shepparton Areas Arrowhead 2,4-D amine is also used in this area to control arrowhead. Channels are sprayed at a rate of 10 L/ha. As in the case of glyphosate, the herbicide is applied with a minimum of water in the channel and where the water is static.

2.2.3. Use of imazapyr in the Murray Valley Area Imazapyr (a.i. 150 g/ L) is used to control arrowhead. Spraying to control arrowhead is usually performed during non-crop season between 15th May and 15th August when the channel is not being used for irrigation. The channel bed is then sprayed with imazapyr at a rate of 5 L/ha using boom spray.

2.2.4. Use of acrolein in Murray Valley Acrolein is rarely used in the Murray Valley at the higher rate of 2.58 mg/L to control general aquatics. Applications are carefully recorded as illustrated in Table 5.

2.3. Channels in the Central Goulburn and RochesterCampaspe Areas The Central-Goulburn and Rochester-Campaspe Areas are centered on the townships of Tatura and Rochester and together comprise 290,103 ha of which 179,816 ha are irrigated. Some 44% of the water of each irrigation area is used for cropping and grazing and a further 15% is used for horticulture (stone fruits, pome fruits and tomatoes). There are 1353 km of channels in the Central Goulburn Area and 599 km in the Rochester Area, making a combined total of 1860 km. In these areas the channel sizes vary from 3 to 5 m in width and from 0.5 to 1 m in depth.

Table 5. Example of detail use of acrolein application Parameter/Variable

Measurement/Condition

Information Source

Ross Gledhill

(Irrigation Area) Channel

MV No.3 @ Offtake from 2 Main

#2482314 File: 2006/1480/1

7

Design Flow rate

350 ML/d

Weed Restricted Flow Rate

200 ML/d

Channel profile

12 – 15 m wide X 1.5 m deep

Weeds treated

(Submerged) Elodea and Aponogeton

Infestation density

Patchy. Some very dense

Time of treatment

Mostly January, but could start in December

Treatment

Acrolein @ 2- 3 ppm, Baker Petrolite method

Length and period of treatment

Target weeds over 10-12 km for 5 hours

Treatment flow rate

200 ML/d

Treatment water conditions



Transport and fate of treated water

      

Very turbulent at injection point. Channel is relatively flat and there are ~7 drop regulators that will increase local turbulence over the length of channel treated; Water temperature 20˚- 30˚ C in Dec, 25˚ - 30˚ C in Jan; Water likely to be turbid, but no measurements available; pH likely to be 6 -8, but no measurements available. Normal water ordering, delivery and regulation; Channel outfalls closed for 48 hours to 20 km downstream of injection point; Mostly pasture, summer cropping, annual pasture; Pasture properties use flood irrigation - 5% of irrigation water for pasture ends up as tail water - 60% of pasture properties have a reuse sump – tail water retained on farm; Off farm tail water most likely will flow to local table drains and drainage depressions – unlikely to reach waterways or wetlands with 10 days

2.3.1. Use of glyphosate in the Central Goulburn and Rochester-Campaspe Areas Arrowhead There are fewer arrowheads in the channels of these areas as compared to other areas with coverage of 0.5% at a rate of 40 L/ha. This means than much less glyphosate is used. Cumbungi To control the cumbungi the only herbicide sprayed is glyphosate, which is sprayed at a rate of 9 L/ha. Cumbungi often occurs in patches in the channels and may cover 0.1% of the channel system. These spots of channel are sprayed by hand gun when the channel is full. While it would be more effective to treat the cumbungi when the water levels are low in the channels, this would cause major disruption to supply and consequently to irrigation schedules, so this is not financially feasible. It is therefore controlled with a spot spray.

#2482314 File: 2006/1480/1

8

2.4. Channels in the Pyramid Boort Area Pyramid Boort Area is centered on the township of Pyramid Hill and comprises 166,215 ha of which 126,400 is irrigated. The area includes 1302 km of channels which vary from 2 to 4 m in width and 0.3 to 0.5 m in depth.

2.4.1. Use of glyphosate in the Pyramid Boort Area Milfoil and Floating pondweed Milfoil and floating pondweed both occur in the Pyramid Boort Area and they may cover up to 50% of the channel area. To control the weed the water level is first lowered and the channel bed is then sprayed with glyphosate at a rate of 20 L/ha using a boom spray. Cumbungi and cane grass Cumbungi in this area also grows in patches and covers 0.1% of channel system. Glyphosate is spot sprayed with a handgun at a rate of 9 L/ha to control the cumbungi and the canegrass.

2.5. Channels in the Torrumbarry Area The Torrumbarry Area is centered on the township of Kerang and comprises 167,000 ha with 150,000 ha irrigation area suitable for irrigation. The area includes 1385 km of channels. Some channels in this area are large, being up to 8 meter in width and 1.5 meter depth of water.

2.5.1. Use of glyphosate in the Torrumbarry Area Water Milfoil and Floating Pond Weed Milfoil and floating pondweed both occur in the Torrumbarry Area and they may cover up to 40% of the channel area before it is considered to be restricting the flow excessively. To control the weed the water level is first lowered and the channel bed is then sprayed with glyphosate at a rate of 20 L/ha using a boom spray. Weeds on the channel banks Weeds on the channel banks and the access pathways are also controlled with glyphosate. This would be applied with a boom spray at a rate of 20 L/ha.

2.5.2. Use of dalapon in the Torrumbarry Area Dalapon (a.i. 740 g / kg) is used to control cumbungi. Spraying is usually performed between 15th May and 15th August when channel is not being used for irrigation. Under these conditions there is no water in the channel except rain water. The channel bed is then sprayed with dalapon at rates of 600 L/ha including 2 kg imazapyr per 100 L using boom spray. #2482314 File: 2006/1480/1

9

2.5.3. Use of acrolein in the Torrumbarry Area Acrolein is rarely used in the Torrumbarry Area. When it is used to control general aquatics it is applied at 0.258 mg/L for 48 hours.

2.6. Drains in the G-MW Irrigation Region The drains in the region are lower than the surrounding paddocks to enable water from the local fields to be fed by gravity. This contrasts with the channels which are typically above the ground level so that water can be fed by gravity onto the fields. The drains serve both to direct floodwaters back to the main river system or, as is more often the case, to remove excess irrigation water. There is therefore potential for them to collect contaminants (including herbicide residues) from the fields. The water depth in drains is much less than in the channels, and typically varies from between 0 - 0.3 m. The drains are usually much narrower than the channels with their width varying between 1 and 2 m although some drains can be as wide as 6 m. A further important difference between drains and channels is that channels are deeper than drains – often the water in a drain is less than 0.1 m deep, and at times the drains may dry completely.

2.6.1. Use of glyphosate in drains Arrowhead Arrowhead is a common weed in drains and retards flowing water. Arrowhead is typically sprayed with glyphosate at a rate of 27 L/ha. General aquatics Glyphosate is also used to control general aquatic plants. To control general aquatics it is applied by rate of 15 L/ha on both the bottom and the sides the drains. Water milfoil and Alisma Water milfoil and Alisma can cover up to 70% of the drain area. These weeds are controlled with glyphosate at the rate of 20 L/ha applied on drain bed.

2.6.2. Use of amitrole in drains Arrowhead and general aquatic weeds Amitrole is used to control general aquatic weeds in drains by spraying at a rate of 11 L/ha (active ingredient of 250 g/L) is used. This gives total 3025 g/ha of the herbicide applied directly over the water in the drain.

#2482314 File: 2006/1480/1

10

3. Environmental fate and toxicological properties of the herbicides 3.1. Glyphosate 3.1.1. Toxicity of glyphosate to different receptors The toxicity of various receptors to glyphosate is given in Table 6. Different endpoints are needed for different receptors. Some, for example tomato, have an EC05 (concentration for 5% reduction) that has been deduced from the literature whereas others, such as the maximum permissible concentration in irrigation water, are set by regulation. From the table it is clear that the lowest tolerance of glyphosate is for irrigation water. Table 6. Toxicity of glyphosate to different receptors Species

Criterion

Tolerance (mg/L)

Reference

Green Algae (Scenedesmus acutus)

EC50

4.00

Saenz et al. 1997

Water flea (Daphnia magna)

Population reduction

1

Bowmer (1987 pp.304)

Fish (Rainbow trout)

LC50

8.0

Tu et al. 2001 (for Roundup)

Aquatic life in fresh water

95% protection

1.2

ANZECC AND ARMCANZ 2000

Mallard duck

NOAEL

4000

Calculated from FAO 2000

Rat

NOAEL

300

Calculated from USEPA 2006

Plant and crop

NOAEL

2.2

Bowmer (1987 pp.304)

Crop (Tomato)

5% reduction

2.7

Calculated from (Santos et al. 2006)

Irrigation Water

Guideline Value

0.1

ANZECC AND ARMCANZ (2000)

Drinking water

Max. permissible conc.

0.7

USEPA (2007)

3.1.2. Degradation in water Glyphosate is readily adsorbed to suspended organic and mineral matter in water and it is then largely unavailable and persistent, since glyphosate is stable to breakdown by sunlight (USEPA 1992). Volatilization or photo-degradation losses are expected to be negligible in most cases. Glyphosate will dissipate rapidly from natural water bodies through adsorption to organic substances and inorganic clays, and dilution. Residues adsorbed to suspended particles will eventually settle into the bottom sediments. There is a range of half lives given for glyphosate in water. The average half-life in pond water given by USEPA is 12 days to 10 weeks (USEPA 1992). From a #2482314 File: 2006/1480/1

11

comparison of different conditions and also report of Layton (2000), a 10 week halflife was used in the calculations.

3.1.3. Degradation in soil Glyphosate is moderately persistent in soil. It is strongly adsorbed to most soils, even those with lower organic and clay content (Wauchope et al. 1992). The major route of glyphosate breakdown in the environment is microbial degradation of soil. Unbound glyphosate molecules are degraded at a steady and relatively rapid rate by soil microbes but bound glyphosate molecules are biologically degraded at a slower rate. Glyphosate half-life values in soil quoted in the literature vary between 1 to 174 days by variation of soil type (Wauchope et al. 1992). Glyphosate in moist silt, clay and loam soil wills disappear in 2 or 3 weeks (EPA 1992). One Australian study has produced an estimate of 28 days for the half-life for local area (Bowmer 1987). This report used an average half-life of 47 days (Wauchope et al. 1992). This is a commonly accepted value (e.g. Layton 2000).

3.1.4. Run-off of glyphosate Because glyphosate binds strongly to soils, it is unlikely to enter waters (even though it is highly soluble in water) through surface or subsurface run-off except when the soil itself is washed away by run-off (Rueppel et al. 1977). Laboratory studies show it does not leach appreciably, and has low potential for run-off (except as adsorbed to colloidal matter) (Wauchope et. al. 1992). Some research shows that the average loss of glyphosate in run-off is unlikely to exceed 2% of the applied chemical (Malik et al. 1996). This is consistent with other experimental data where the maximum run-off from an experimental study was reported as 1.85% of the applied glyphosate (Cheng et al. 1990). A value of 2% was therefore taken as the reasonable maximum estimate of run-off for glyphosate.

3.1.5. Degradation of glyphosate in plants Glyphosate is quickly absorbed by leaves and roots of plants. Once absorbed into the leaves, glyphosate is broken down slowly. It moves quickly through the plant and accumulates in areas of active growth. Spraying a plant with glyphosate inhibits protein and amino acid synthesis in that plant. This lack of amino acids stops plants growing and within a week or so, the plant tissues and organs slowly degrade due to lack of proteins. Death of the weed ultimately results from lack of nutrients and dehydration occurs a week or so later (Ross and Childs 2007). Metabolic degradation of glyphosate in plant is disputed. Some scientists believe that glyphosate is not metabolized by plants (Schuette 1998), while some other researchers claim that some plants are able to metabolise glyphosate (Carlisle and Trevors 1988). Different plants can apparently degrade glyphosate at different rates, and weeds are able to degrade glyphosate in a shorter period. The half-life estimate of 2.5 days as used by Layton (2000) has taken this into account in this report. The half-life of glyphosate on foliage has been estimated at 10.4 to 26.6 days (Newton et al. 1984).

#2482314 File: 2006/1480/1

12

3.2. Amitrole 3.2.1. Toxicity of amitrole to different receptors The toxicity of amitrole to different receptors is shown in Table 7. As shown in the table, the maximum permissible concentrations for irrigation and human health are very low (0.001-0.002 mg/L) Table 7 Toxicity of amitrole to different receptors Receptor

Tolerance criterion

Concentration (mg/L)

Reference

Green algae (Pseudokirchneriella subcapitat)

EC50

1

Wang et al. (1990)

Aquatic Plant

EC50

2.5

Wolf (2001)

Water flea (Daphnia magna)

Reproduction

0.2

Ritter (1989)

Fish (Rainbow trout)

NOAEC

100

Wolf (2001)

Fish

Reproduction

0.2

Abbott (1994)

Honeybee

NOAEL

100

Wolf (2001)

Quail, duck

Reproduction

100

Wolf (2001)

Rat

NOAEL

10

Calculated from Weber (1978)

Dog

NOAEL

62.5

Calculated from Weber (1978 )

Irrigation water

Guideline value

0.002

ANZECC AND ARMCANZ (2000)

Drinking water

Guideline value

0.001

ANZECC AND ARMCANZ (2000)

3.2.2. Degradation of amitrole in water Amitrole is only expected to breakdown slowly by hydrolysis or photolysis in an aquatic environment, with a reported half-life of 40 days and the half-life is even longer in pond water (Howard 1989). Degradation of amitrole in open water may occur through oxidation by other chemicals. Amitrole does not volatilize because of its low vapour pressure and it will be remain in water due to its high solubility. The main route of removal from water may be through adsorption to sediment particles. On the other hand with photo-degradation in the presence of the photosensitizers (e.g. humic acid and potassium salt) the half-life will decrease to several hours (Abbott 1994). The half-life estimate of 28 days (as used by Layton, 2000) was considered more appropriate for this study.

#2482314 File: 2006/1480/1

13

3.2.3. Degradation of amitrole in soil Loss of amitrole from soils by volatilization or photo-degradation is minor. Some chemical degradation may occur in soil (Abbott 1994). Amitrole does not adsorb strongly to soil particles and it is readily soluble in water, and it therefore has a moderate potential for groundwater contamination (PMEP 1984). Thin-layer and thick-layer chromatography, molecular topology, water solubility and octanol-water partition coefficient (Kow) all predict that amitrole will be easily leached in soil. Amitrole has high mobility in soils of pH > 5 and medium to high mobility in soils with lower pH. Generally, movement is most readily seen in sands and increased organic matter content reduces mobility (Abbott 1994). The soil dissipation rate of amitrole is affected by moisture, temperature, cation exchange capacity, and clay content, but is unaffected by soil pH. The half-life of amitrole in very low temperature and clay soil is quite long (100 days) (PMEP 1984), while in warm soil the period decreases to 14 - 21 days (WSSA Herbicide Handbook Committee 1989). Amitrole is stable to degradation from abiotic hydrolysis and aqueous photolysis, and is slightly to moderately persistent (aerobic soil metabolism half-life 22 - 26 days; aerobic aquatic metabolism half-life 57 days) in aerobic environments. The generally accepted value for the half-life of amitrole is 14 days, and this has been used in this report.

3.2.4. Run-off of amitrole Because amitrole does not sorb significantly onto soil particles, it may be transported in the dissolved phase by run-off to surface water bodies. Amitrole may contaminate surface water from run-off or spray drift associated with ground spray application.

3.2.5. Degradation of amitrole in plants The metabolic pathways of amitrole in plants appear to be complex. There is evidence that when amitrole is applied to the leaves of plants, most of the material absorbed is metabolized (INCHEM 1974). Research has shown that this process takes from 1 to 4 weeks (Weed Science Society of America 1994). A shorter half-life of 4 days has been reported for amitrole applied to cotton (Miller and Hall 1961). A half-life of 5 days for amitrole in weeds has been used in this report.

3.3. 2,4-D amine 3.3.1. Toxicity of 2,4-D amine to different receptors The toxicity of 2,4-D amine to a range of receptors is given in Table 8. Similar to amitrole, its maximum acceptable concentration for irrigation and drinking water use is quite low, but higher than amitrole. Table 8. Toxicity of 2,4-D amine to different receptors Receptor

Tolerance criterion

Concentration (mg/L)

Reference

Algae

NOAEC

26.4

Hughes et al. (1990)

(Selenastrum capricornutum)

#2482314 File: 2006/1480/1

14

Duckweed (Lemna gibba)

NOAEC

2.029

Hughes et al. (1997)

Aquatic Invertebrates

NOAEC

19.7

INCHEM (1997)

Fish (Rainbow trout)

NOAEC

0.0164

Xie and Thrippleton (2005)

Aquatic life in fresh water

95% protection

0.28

ANZEEC and ARMCANZ (2000)

Rat

NOAEL

3.41

Calculated from (Fagliari et al. 2005)

Crop (Tomato)

No damage

0.15

Calculated from (Fagliari et al. 2005)

Crop (Soybeans)

Damage LOEL

0.22

Que et al. 1981

Irrigation water

Guideline value

0.03

ANZECC and ARMCANZ (2000)

Drinking water

Max. permissible conc.

0.07

USEPA (2006)

3.3.2. Degradation of 2,4-D in water In water with a low pH, 2,4-D will remain in a neutral molecular form, increasing its potential for adsorption to organic particles suspended in water, and this increases its persistence. Adsorption also increases in muddy water. Microorganisms readily degrade 2,4-D in the aquatic environment, with the rate increasing with increasing nutrients and dissolved organic carbon (EXTOXNET 1996). The half-life of 2,4-D amine in an anaerobic aquatic environment can be as long as a year, but this time will significantly decrease to 15 days in aerobic aquatic environment (USEPA 2005). In natural water a range of 4 to 28 days has been estimated by the USEPA (2006). A half-life of 20 days was chosen for this study.

3.3.3. Adsorption and degradation of 2,4-D in soil Soil organic content and soil pH are the main determinants of 2,4-D adsorption in soils. Adsorption increases with increasing soil organic content and decreasing soil pH (Johnson et al. 1995). 2,4-D degradation rates in soils degrade at the same rate with and without sunlight, indicating that photo-degradation is not an important process in the field (Johnson et al. 1995). This suggests that soil microbes are primarily responsible for 2,4-D amine’s disappearance (Howard 1989). Degradation rates are determined by the microbial population, environmental pH, soil moisture, and temperature. A number of microbial organisms rapidly degrade 2,4-D. In sediments with a sufficient microbial population, 2,4-D can be degraded in a matter of hours (Aly and Faust 1964). A range of halflives of 1.25 h to 40 days are present in the literature but most estimates are between 3 and 10 days. However, in cold dry soil the half-life is longer (FAO 1997). In this study, a half-life of 10 days has been used.

#2482314 File: 2006/1480/1

15

3.3.4. Run-off of 2,4-D The rapid biodegradation of 2,4-D in soil prevents significant downward movement under normal field conditions. Run-off from treated soil has been estimated at between 0.01 and 1% of the applied 2,4-D (FAO 1997).

3.3.5. Degradation of 2,4-D in plants 2,4-D is a plant hormone (auxin) mimic. It causes rapid cell division and abnormal growth. Absorbed 2,4-D by foliage or roots tends to accumulate in growing tips. Metabolism of 2,4-D in plants is by a variety of biological and chemical pathways (Herbicide Handbook 1994) but it is generally slow. 2,4-D amine can usually remain active against susceptible plants for 1 to 4 weeks (Wilson et al. 1997). In one study the half-life of applied 2,4-D on grass was estimated to be 14 days. GLEAMS model (Layton, 2000) uses a half-life of 9 days for 2,4-D amine in weeds. That half-life has been used in this study.

3.4. Dalapon (2,2 DPA) 3.4.1. Toxicity of dalapon to different receptors The toxicity of various receptors to dalapon is given in Table 9. Different endpoints are needed for different receptors; for instance, an EC50, NOAEC and NOAEL are deduced from the literatures while, others such as the maximum concentration level for drinking water is set by regulation.

3.4.2. Adsorption and breakdown in soil Dalapon does not readily bind or adsorb to soil particles. Even in soils with very high organic levels, as little as 20% of applied dalapon may be adsorbed. In clay and clay loam soils, there may be no adsorption (Howard 1989). Since some micro-organisms (bacteria) can use dalapon as an energy source, dalapon absorbed into soil disappears rapidly via their activities. Biological breakdown, also referred to as biodegradation, is the main route of dalapon disappearance from soils. Dalapon was not detected below the first six-inch soil layer by Doyle (1984). Degradation depends on soil type, temperature, and moisture. Soils that were buffered to a pH of 6.5 provided the best conditions for the adaptation of microbes and degradation of the herbicide. At higher temperatures, dalapon can also be degraded by ultraviolet light from the sun (photodegradation). Generally higher temperatures and increased soil moisture speed up the rate of degradation (Doyle 1984). A half-life of 30 days is used in this study.

3.4.3. Run-off Since dalapon does not adsorb to soil particles, it has a high degree of mobility in all soil types thus, leaching and run-off do occur. However, dalapon movement in soil is usually limited by rapid and complete breakdown of the herbicide by soil microorganisms into naturally-occurring compounds (Howard 1989). Furthermore, in a #2482314 File: 2006/1480/1

16

recently conducted national groundwater survey, dalapon was not found in groundwater as a result of its agricultural use (Cohen 1984), although it has been detected in surface water run-off (Bellingham 2005). Because of the low Koc we assume that up to 5% of the applied dalapon applied to the stream bank could potentially run-off into the channel under favourable conditions. Table 9. Ecotoxicological values for different receptors for dalapon Receptors

Criterion

Tolerance

References

Blue-Green algae (Anabaena sp.)

Mortality

10 mg/L

Venkataraman and Rajyalakshmi (1972)

Duckweed (Lemna minor)

Physiology general

1.43 mg/L

O'Brien and Prendeville (1979)

Brown shrimp

EC50

1 mg/L

WSSA (1989)

Crustaceans

NOAEL

200 mg/L

WSSA (1989)

Gold fish

No death

100 mg/L

Meister (1992)

Bluegills

LC50

105 mg/L

Meister (1992)

Chicken

LD50

500 mg/L

Hartley and Kidd (1983)

Rat

NOAEL

37.5 mg/L

WSSA (1989)

Dog

NOAEL

1000 mg/L

WSSA (1989)

Crop (Beets)

Injury threshold

7.0 mg/L

ANZECC and ARMCANZ (2000)

Crop (Corn)

Injury threshold

0.35 mg/L

ANZECC and ARMCANZ (2000)

Crop (Grapes)

NOAEC

5 mg/L

Leonard et al. (1964)

Irrigation water

Guideline value

0.004 mg/L

ANZECC and ARMCANZ (2000)

Drinking water

Max. conc.

0.2 mg/L

USEPA referenced by Rice (1997)

permissible

3.4.4. Adsorption and breakdown in water Dalapon and its known breakdown products dissolve easily in water (EXTOXNET 1996). If dalapon enters to water bodies such as ponds or streams, it is removed via microbial degradation, hydrolysis and photolysis (Howard 1989). Microbial degradation tends to be the most active form of its breakdown in water (US DOA 1984). Under conditions favourable for microbial growth, dalapon decomposition via microorganisms will probably be completed within one month which will diminish the importance of chemical hydrolysis (EPA Ground Water and Drinking Water). In the absence of microbial degradation, the half-life of dalapon, by chemical hydrolysis, is #2482314 File: 2006/1480/1

17

several months at temperatures less than 77 degrees F (25 C) (US DOA 1984). Hydrolysis increases with increase in temperature and pH (e.g., in alkaline waters). Direct photolysis in water may be possible, although photolytic rates have not been investigated under environmental conditions. In addition, aquatic volatilization and adsorption to sediments are not expected to be significant (EPA Ground Water and Drinking Water). Following the EPA Ground Water and Drinking Water study, a halflife of 30 days in water has been used in this study.

3.4.5. Absorption and breakdown in plants Dalapon is absorbed by plant roots and leaves and translocated within plants (Worthing 1983). Translocation occurs from the leaves to the roots of most species given foliar treatment with the herbicide (EXTOXNET 1996). At high rates of application dalapon 'precipitates' out of solution as an acid, and has immediate and local acute effects on foliage (US DOA 1984). It is easily washed off foliage. Conditions of increased light and high temperature may cause nutrient solutions or soil applications of dalapon to build up in the tops of plants, via transpiration (EXTOXNET 1996). Dalapon-sodium is persistent in plants (Hartley and Kidd 1983). It accumulates in young tissue and is not degraded rapidly (WSSA 1989). GLEAMS (Layton, 2000) uses an estimated half-life of 37 days in foliage, and we have accepted that value in this study.

3.5. Imazapyr 3.5.1. Toxicity of imazapyr to different receptors The toxicity of various receptors to imazapyr is given in Table 10. Different endpoints have been reported for different receptors; for instance, 15% mortality of corn and soybean due to imazapyr exposure has been reported by Shaner and O’Connor (1991) could be called an EC15 while, NOAEC value has been collected for rat and Northern bobwhite quail. Table 10. Toxicological values of imazapyr for different receptors Receptors

Criterion

Tolerance

References

Green algae (Chlorella emersonii)

EC25 Growth

0.02 mg/L

Landstein et al. 1993

Duckweed (Lemna gibba)

EC25 Growth

0.0132 mg/L

Mangels and Ritter 2000

Water flea (Daphnia sp.)

NOAEC

40.7 mg/L

Cyanamid 1997

Pink shrimp

NOAEC

132 mg/L

Mangels and Ritter 2000

Benthic macro invertebrate

NOAEC

18.4 mg/L

Fowlkes et al. 2003

LC50

>100 mg/L

WSSA 2004

Fish (Rainbow sunfish)

#2482314 File: 2006/1480/1

trout,

bluegill

18

Northern bobwhite quail

NOAEL

200 mg/L

Pless 2005

Rat

NOAEL

10,000 mg/kg

Shaner and O’Connor 1991

Dog

NOAEL

10,000 mg/kg

Shaner and O’Connor 1991

Rabbit

NOAEL

400 mg/kg

Shaner and O’Connor 1991

Crop (Corn)

*EC15

0.18 mg/L

Shaner and O’Connor 1991

Crop (Soybean)

EC15

0.9 mg/L

Shaner and O’Connor 1991

* 15% mortality

3.5.2. Adsorption and breakdown in soil The adsorption of imazapyr to soil particles is generally weak and it varies depending on soil properties (Mangels 1991). Adsorption is reversible, and desorption occurs readily (WSSA 1994). Because the chemical form of the herbicide is determined by environmental pH, the adsorption capacity of imazapyr changes with soil pH. Below pH 5 the adsorption capacity of imazapyr increases and limits its movement in soil. Above pH 5, greater concentrations of imazapyr become negatively charged, and fail to bind tightly with soils (Mangels 1991). There is a partitioned between organic carbon and water, resulting in imazapyr being available for plant uptake (Tu 2001). Vizantinopoulos and Lolos (1994) found that adsorption decreased with increasing soil temperature, and Dickens and Wehtje (1986) found that adsorption increased with time and decreased soil moisture. In general, imidazolinone herbicides show an increase in soil adsorption capacity with an increase in soil clay content and organic matter (Tu 2001). In soils, imazapyr is degraded primarily by microbial metabolism. It is not, however, degraded significantly by photolysis or other chemical reactions. The half-life depends on environmental circumstances factors affecting degradation rates were difficult to identify because the pH varied with temperature and organic content (McDowell et al. 1997). The half-life of imazapyr in soil ranges from one to five months (Tu 2001). Ranges of half-life of imazapyr in soil vary from 25 to 155 days depending on conditions; the half-life of 90 days estimated by Layton (2000) is taken in this study.

3.5.3. Run-off Under most field conditions imazapyr does not bind strongly to soils and can be highly mobile in the environment. Above pH 5, the herbicide will take on anionic form, increasing the risk of herbicide run-off. McDowell and his colleagues in 1997 found that heavy rainfall caused significant movement of the herbicide and leaching up to 50 cm deep in soils have been reported (WSSA 1994). Despite its potential mobility, Tu (2001) claimed imazapyr has not been reported in water run-off in any field studies. However, imazapyr has been detected in shallow groundwater in Yellowstone County (Anon. 2005) and Cox (1996) gives examples of detections in both surface and groundwater. In the current case we have assumed that 2% of the imazapyr that was applied to the banks runs off into the channel.

#2482314 File: 2006/1480/1

19

3.5.4. Adsorption and breakdown in water Imazapyr with water solubility of 11,000 mg/L is easily dissolved in water. The primary form of degradation in water is photo-degradation with a half-life of approximately 2 days (Mallipudi et al. 1991). Microbial degradation can accelerate the degradation of imazapyr. Degradation decreases with increasing pH. Due to its rapid photo-degradation by sunlight, water contamination by imazapyr is generally not of concern. Degradation via volatilization of imazapyr is reported to be minimal in field studies (Hgonzales 2004). Considering environmental circumstances, the half-life of 4 days in water as reported by Hgonzales (2004) has been used in this study.

3.5.5. Absorption and breakdown in plants Imazapyr is absorbed rapidly through plant tissue and can be taken up by roots. It is translocated in the xylem and phloem to the meristematic tissues, where it inhibits the enzyme acetohydroxy acid synthase (AHAS), also known as acetolactate synthase (ALS). The rate of plant death usually is slow (several weeks) and is likely to be related to the amount of stored amino acids available to the plant (Tu 2001). Shaner and O’Connor 1991 have studied the translocation of imazapyr in corn. After 3 days 18.3% of the herbicide was on treated leaves, while just 12.5% of herbicide remained in the leaf and the rest had translocated to roots. Translocation in some plants takes a few days; for instance, in one study translocation of imazapyr in Imperata cylindrica plant reached a maximum of 60% after 9 days. Conditions such as low humidity, winds and high temperature can decrease the total amount of herbicide that is absorbed by foliage (Shaner and O’Connor 1991). Metabolism of imazapyr varies from plant to plant. Soybean can metabolize 20% of absorbed imazapyr in 2 days while a longer half-life has been determined for other plants (Shaner and O’Connor, 1991). Layton (2000) uses an estimated half-life of 30 days in foliage and that value has been used in this study.

3.6. Acrolein 3.6.1. Toxicity of acrolein to different receptors The toxicity of acrolein to different receptors is given in Table 11 together with the sources of that information. Table 11. Toxicity of Acrolein to different receptors Species

Tolerance criterion

Toxic Conc. (mg/L)

Reference

Algae (Selenastrum capricorutum)

EC50

0.00005

Tomlin (2000)

Duckweed (Lemna gibba)

EC50

0.07

Tomlin (2000)

Water flea (Daphnia sp.)

LC50

0.022

Siemering et al. (2005)

Goldfish

NOAEC

0.0114

Bridie et al. (1979)

Aquatic life

Safe level

0.001

Victorian EPA (2006)

#2482314 File: 2006/1480/1

20

Rabbit

NOAEL

0.05

Calculated from USEPA (2000)

Dog

NOAEL

1

Calculated from USEPA (2000)

Crop (Soybean)

NOAEC

15

USEPA (1973)

Crops and pasture

Max. permissible conc.

1.5

USEPA (1980) Ambient Water Quality criteria for acrolein

Drinking water

Max. permissible conc

0.32

USEPA (1987)

3.6.2. Adsorption and break down of acrolein in water Acrolein has high solubility in water and low Koc and therefore if released into water is not expected to adsorb to suspended solids and sediments (HSDB 2003). Generally acrolein has a short half-life in water in the field. The half-life in water depends on water temperature, turbidity, weed load, oxygen concentration, volatilization (due to high vapour pressure) and also the influence of micro-organisms (Bowmer and Sainty 1991). The rate of reaction of acrolein increases with increasing pH. In flowing water, the rate of loss was much faster reflecting the influence of turbulence in increasing loss through volatilization (Hutson and Roberts 1987, pp300). USEPA’s toxicological review on acrolein cites a half-life of 4.4 hours in a model river (HSDB 2003). USEPA calculated half-lives of acrolein from degradation rate constants in irrigation canals to be in the range of 3 - 7 hours (USEPA 2003). Other half lives reported in literature in irrigation channels range from 4 - 10 hours. The most relevant data are the Australian studies by Bowmer and colleagues reviewed by Bowmer and Sainty (1991) on dissipation of acrolein from irrigation channels under different flow conditions and temperature. That review showed the half-lives ranging from 3.3 to 6.7 hours. Most half-life data in literature also falls within 3.3 to 10.2 hours. For this study we have used 5 hours for the half-life of acrolein in channels.

3.6.3. Degradation of acrolein in plants Biochemical and toxic effects of acrolein are probably caused by its reaction with critical protein and non-protein sulfhydryl groups (USEPA 2003). The reaction of acrolein with sulfhydryl compounds is rapid and essentially irreversible, resulting in the formation of stable thiol ether. When added to water as an aquatic herbicide, acrolein undergoes rapid decomposition, especially in the sunlight. At the same time, it reacts rapidly with amines, alcohols, and mercaptans of aquatic plants, destroying cell structure and killing the plants (USEPA 2003).

#2482314 File: 2006/1480/1

21

3.7. Summary of pesticide properties A summary of the physical and chemical properties of the six pesticides is shown in Table 12.

References

Acrolein

Imazapyr

Dalapon

2,4-D amine

Amitrole

Properties

Glyphosate

Table 12 Properties of herbicides used by G-MW and considered in this report

Water solubility (mg/L)

12,000

280,000

900

900,000

11,000

208,000

Bowmer (1987)

Log Kow (octanol: water partition coefficient)

< -3.2

-0.97

2.58

0.77

1.3

1.08

Tomlin (2000)

Koc (adsorption coefficient)

24,000

100

20

1

100

0.5

Layton 2000

Pesticide mobility rating

Low

Moderate

Moderate

High

High

Very High

Vogue et al. 1994

Volatilization

No

No

No

No

No

Yes

Bowmer (1987)

Vapour pressure (mpa)

0.00131

1?

Yes Tier 2

Figure 4 Diagrammatic steps involved in a Tier 1 assessment http://www.epa.gov/oppefed1/ecorisk/setac98b.pdf

4.3. Tier 1 Estimation of PEC The maximum application rate was assumed (expressed as L/ha) in each case where a herbicide was applied to a channel or drain. The given application rates of herbicides were multiplied by the fraction of active ingredient (a.i.) and converted to the active rate of application. The PEC was calculated for each herbicide using different assumptions involving channel depth or drain depth together with the application rates for each scenario.

4.3.1. Calculation process of PEC for herbicide applied directly to a channel or drain In many situations, weeds in a channel or drain are controlled by a boom spray directed at the weeds emerging through the surface of the water in a channel or a drain. Under those conditions it is possible to use a simple calculation to estimate #2482314 File: 2006/1480/1

25

the PEC. The steps used for such a calculation are given below.

10% run-off

10% run-off

1m

1m

2m

Figure 5. Diagrammatic cross-section of channel

One scenario is when the channel and banks have been sprayed, and then some run-off occurs through rain before the channel is refilled (Figure 5). In that case there will be a contribution to the pesticide load in the channel from the bank. When the channel is refilled, the volume of water will be determined by the channel volume. Inclusion of the spraying of the banks can therefore potentially increase the amount of pesticide while the volume of water is fixed, leading to an increase in the concentration in the channel. When the banks are sprayed, the length of channel per hectare sprayed area could be as shown in Figure 6.

a. Unsprayed Banks

1m Channel 5000m

2m

1m

b. Sprayed Banks

1m Channel 2500m

2m

1m

Figure 6. Comparisons of when bank is sprayed or not sprayed. Note, in both cases the area sprayed is one hectare.

#2482314 File: 2006/1480/1

26

An alternative scenario is that the rain occurs after the channel has been filled. In that case the water in the channel could be diluted by the water from the bank. The previous scenario, when the rain occurs before the channel is filled, is therefore the worst case scenario. That scenario is therefore considered in the Tier I risk assessment as shown in Table 14. Table 14 Calculation of PEC, when water area is sprayed only, or banks and water area both are sprayed at a rate of 40 L/ha with 0.36 a.i. Case

Banks not sprayed

Banks sprayed

Area sprayed

1 ha

1 ha

Application rate

14.4 kg a.i. / ha

14.4 kg a.i. / ha

Sprayed amount over water

14.4 kg a.i.

7.2 kg a.i.

Sprayed on to bank

0 kg a.i.

7.2 kg a.i.

Total sprayed

14.4 kg a.i.

14.4 kg a.i.

Fraction run-off from bank

0.1 = 10%

0.1 = 10%

Contribution from bank

0 kg a.i.

0.1 × 7.2 kg = 0.72 kg a.i.

Potential amount going to water

14.4 kg

7.9 kg

Assumed depth

0.4

0.4

Waterway width

2m

2m

Bank width sprayed

0m

1 m on each side

Total width sprayed

2m

4m

Channel length for 1 ha

5000 m

2500 m

Water needed to fill channel

5000 m × 2 m × 0.4 m = 4 ML

2500 m × 2 m × 0.4 m = 2 ML

Potential concentration

14.4 kg / 4 ML = 3.6 mg/L

7.9 kg / 2 ML = 4.0 mg/L

For a 2 m wide drain or channel, using the same logic as in Table 14, the maximum predicted concentration is: PEC = (1.1 × A × a.i. /d /10) mg/L

Equation 1

where A is the application rate in kg/ha, a.i. is the fraction of active ingredient, and d is the depth of the channel or drain in m.

4.3.2. Estimation of PEC for Acrolein The application of acrolein differs from most other herbicides in that it is injected into the water rather than being applied to the plant surfaces. Nowadays, acrolein is seldom used in irrigation channels, and when it is applied the application rate does not exceed label rates of 15 mg/L. G-MW applies acrolein at #2482314 File: 2006/1480/1

27

two rates, namely 3 and 0.3 ppm (mg/L) measured on a volumetric scale. The lower rate involves a longer application time. Note that the rates used by G-MW are actually much less than the approved label rates. Acrolein with specific gravity of 0.862 gives the concentrations in water as shown in Table 15. Table 15 Conversion of volumetric to gravimetric concentration for acrolein Volumetric concentration

Gravimetric concentration

3 µL/L = 3 ppm

2.58 mg/L

0.3 µL/L = 0.3 ppm

0.258 mg/L

In this Tier 1 assessment, in common with our assessment of the previous herbicide, no losses are considered and the entire applied amount is assumed to be mixed in channel water. The Tier 1 PEC is therefore 2.58 mg/L for the higher application rate and 0.258 mg/L for the lower rate. The higher rate is used in the Murray Valley Irrigation Area and the lower rate (but with 48 hour exposure) is used in the Torrumbarry Area.

4.4. Tier 2 Estimation of PEC Problem definitions are the same in Tier 1 and 2 (Figure 7). However, in Tier 2, the estimation of the PEC considers various processes and pathways of loss. These included plant interception, dilution, volatilization, adsorption, biodegradation and effective half-life of herbicide in each compartment (e.g. water, soil and plant), an estimate of the fraction that would run-off from the bank and the fraction of herbicide applied to a plant that would have washed off from the plant after a channel was refilled with water. This is illustrated in Figure 8. In the case of acrolein, a different method is used for PEC calculation. In this case dispersion and diffusion models from literature were adapted to predict PEC. The more accurate Tier 2 PEC was then used to estimate a HQ. No safety factor was included at this stage, although typically near worst conditions were considered (e.g. channels with the shallowest depth of water was used).

#2482314 File: 2006/1480/1

28

Problem formulation

Hazard

Effects on receptors

Herbicide exposure (More realistic assessment involving dissipation, dilution , biological breakdown, etc )

Quotient

No No more assessment

HQ>1? Yes Risk

management recommendation

More data are required

Figure 7. Diagrammatic steps involved in a Tier 2 risk assessments From http://www.epa.gov/oppefed1/ecorisk/setac98b.pdf

Spray over channel or drain

Direct to water (10-40%)

Intercepted by plant (60-80%)

Intercepted by soil (0-10%)

Break down in plant

Break down in soil

Break down in water

Absorbed into bottom sludge (Koc)

Washed off into Water

0-2 % run-off From Embankment

Back to water due to equilibrium (Kd)

Stay in water (mg/L)

PEC (mg/L)

Figure 8. Method for Tier-2 assessment of the concentration of pesticide in water following spraying with herbicide. Note that the calculations of PEC involved the application rate but are not influenced by the area sprayed. #2482314 File: 2006/1480/1

29

4.4.1. Processes considered in Tier 2 As stated earlier, in Tier 2 the factors which affect final herbicide concentration present in water are described and used in the calculation of PEC. Processes such as biodegradation and effective half-life of herbicide in each compartment (e.g. water, soil and plant), fraction run-off and wash-off from plant after application time are considered. How these processes are incorporated is described below. Typically the water level in a channel is lowered to between 5 and 15 cm before spraying. Also no herbicides are applied when rain is forecast – typically this means there would be at least 4 days between the time of spraying and rain. Spraying for weeds using a boom spray typically would occur only when the weed cover exceeds 50%, although spot spraying with a hand gun may occur when there is less weed cover. Spot spraying uses a small fraction of the amount of herbicide (perhaps 1%) compared to a boom spray, so the effort has been focussed on the boom spray as that presents a larger risk. Of the herbicide applied by a boom spray, 50% to 80% would have been intercepted by foliage and only 20% - 50% would be directly mixed into water and a small fraction (0 -10%) could be sprayed directly on to soil if the water level was very low. In addition, some spray would be applied to the bank (Figure 9). Each of these four pathways is discussed below.

Figure 9. Water status in channels and drains at the time of herbicide application.

The plant cover is unlikely to be complete, so some herbicide will be intercepted by water. The rate of degradation of herbicides depends on several factors. Some herbicides with higher vapour pressure volatilize quickly (refer to Table 12) and this process would be dependent on the water temperature, water depth, turbulence and #2482314 File: 2006/1480/1

30

the rate of air movement across the water surface. Another process is photolysis where the herbicide is broken down by light. Often the most important process is biodegradation whereby micro-organisms can use the compound as carbon and energy source. The combination of these methods of pesticide degradation is typically expressed as a half-life or degradation rate. The amount of herbicide that is remaining in the water after t days was estimated using a first order decay model as: Rwater = A × a.i. × Iwater × exp (- kwater × t)

Equation 2

Where A is the application rate (kg/ha), Iwater the fraction sprayed straight on to water and kwater is the breakdown rate in water. The residual pesticide in the water is a component of the pesticide burden of the system.

4.4.2. Interception of herbicides by plants Generally once the herbicide has been applied it will be absorbed by the plant within a few hours. Research shows that herbicide's properties such as the octanol-water partition coefficient (Kow) have an important role in uptake amount of herbicide by plant (Briggs et al. 1982). A large proportion of the herbicide that is intercepted by foliage is translocated through the plant. Some data are available on the degradation rate of herbicides in plants (Table 12), and the amount remaining after a given time can therefore be estimated assuming a first order decay model. Since in this study the amount of herbicide which is sprayed was known, the only other parameters needed are foliage interception and half-life in the plant. The residual amount in the plant (Rplant) after t days is then estimated as: Rplant = A × a.i. × Iplant × exp(-t × kplant)

Equation 3

where A is the application rate, Iplant is the fraction intercepted by plants and kplant is the breakdown rate of the pesticide in the plant.

4.4.3. Wash-off of herbicides from plants When the channel is refilled after spraying (which would be at least 4 days after spraying), a fraction of the residual spray in the plants could be washed off. The fraction that is washed off (W off) is related to a number of factors including the nature of the leaf surface, plant morphology, pesticide solubility, and polarity of the pesticide molecule, formulation of the commercial product, and timing and volume of rainfall. Values of the wash-off factor provided in Layton (2000) for organochlorine, organophosphorus, carbamate, and pyrethroid insecticides are based primarily on the work of Willis et al. (1980), or computed from the algorithms provided relating #2482314 File: 2006/1480/1

31

wash-off to rainfall volume and pesticide solubility. For other pesticides, Koc was used as a guide for estimating W off. Because of the lack of data on W offs, a range of values was used in the calculations. Because of this paucity of data and its sensitivity in the equations used to obtain PECs, we recommend that further data be obtained on this component.

Recommendation: Accurate data needs to be obtained under the conditions relevant to this study for the wash-off fractions, especially for those pesticides showing potential risk under the assessed scenarios The contribution to the waterbody from spray intercepted by plants is: Cplant = W off × Rplant

Equation 4

where Cplant is the contribution to the pesticide burden from wash-off of plants.

4.4.4. Herbicide sprayed onto soil A fraction of the herbicide may be sprayed directly on to soil if the water level is low. The contribution to the water body from spray intercepted by plants is: Rsoil = A × a.i. × Isoil × exp(-t × ksoil)

Equation 5

where Rsoil is the contribution to the pesticide burden from pesticide intercepted directly to soil and is the breakdown rate of that pesticide in soil. The residual from the soil also contributed to the pesticide burden.

4.4.5. Run-off from the bank The rationale for this component has been discussed in Section 4.3.1 and is not repeated here. In keeping with this being a Tier 2 estimate, a more refined estimate of the run-off is made. The first refinement is the fraction of pesticide that is applied to the bank that would run-off into the channel. Rather than using the very high estimate of 10% as was used in Section 4.3.1, a more realistic value was sought. Experience from other situations (Leonard 1990; FAO 1997) suggests the fraction is low. The experience shows that in many case is it < 2% (Cheng et al. 1990) – in this report we have used a slightly higher run-off factor of 2%. This value would vary with the herbicide used. A second refinement was to allow for breakdown of the pesticide between the time of spraying and water entering the channel or drain. The modelling is sensitive to the total time between spraying and the filling of the channel or drain. In the Tier 2 calculations, the same spaying rates and channel dimensions as used in #2482314 File: 2006/1480/1

32

Tier 1 were assumed – namely the channel was 2 m wide with an average depth of 2 m and that the bank was sprayed to a width of 1 m. The contribution of the run-off to the pesticide burden Crun-off is formalised in the following equations: Crun-off = A × a.i. × Frun-off × exp( ksoil × t),

Equation 6

where Frun-off is the fraction of run-off and ksoil is the rate of breakdown of pesticide in the soil.

4.4.6. Mass of soil An estimate of the mass of soil at the bottom of the channel or drain is required for subsequent calculations. There are a range of soil types throughout the G-MW Area. There are few data available on the soil layer at the bottom of the channels or drains but it would be expected that there would be a thin layer that was rich in organic matter (a ‘schlick’) that would be formed on the surface. The thickness of this layer of soil in the channels and drains is assumed to be 10 mm, but is likely to be much thicker. This layer, which is rich in organic matter and fine particles, can sorb pesticides such as glyphosate. The mass of soil affected was estimated by assuming an incorporation depth of 10 mm and a soil bulk density of 1.33 t/m3. From these assumptions we can work out the mass of soil affected for each hectare sprayed would be: 10,000 m2/ha × 1/1000 (mm/m) × 10 mm ×1.33 t/m3 = 133 t/ha The amount of the organic matter present on the bottom of the channel (or drain) is critical to these calculations, but there is little if any data available. We therefore recommend that thickness and organic matter content of the lining layer of the channels (or drains) be measured.

4.4.7. Total pesticide burden The total pesticide burden in the channel was obtained by summing the contributions from the spray falling directly on the water, on the leaves, on the soil at the bottom of the drain or channel and a potential contribution from the bank. These four components were added together and represent the total pesticide burden from that spraying.

4.4.8. Sorption of pesticides by soil Some herbicides are quickly sorbed by the soil particles or organic matter. Soil type plays an important role in sorption of herbicides. The strength of sorption of a pesticide to a soil basically depends on that pesticide’s soil sorption coefficient (Kd) which in turn is a function of the soil’s organic carbon sorption coefficient (Koc) for most pesticides. Glyphosate is an exception, where clay plays an important role. There is a high correlation between the organic matter content of the soil and Kd. This is because the soil organic matter acts as a non-polar phase or surface, which is the main sorbent in soils; this attracts pesticides because they are typically non#2482314 File: 2006/1480/1

33

polar organic molecules. The soil organic carbon sorption coefficient of a pesticide is often calculated by following equation: Koc= Kd/ Foc

Equation 7

where Foc is organic matter percentages in soil (Wauchope et al. 2001). After absorption by soil, the herbicide is mainly degraded by micro-organisms in the soil; meanwhile, due to equilibrium between soil and water phases, some fraction of the herbicide could also be released into water. The degradation rate of the herbicide depends on parameters such as temperature, pH, moisture, caution exchange capacity and clay content (PMEP 1984). The amount of herbicide which is released into water after interception by soil will be calculated by:

Ps 

1  Kd   V  ms   1  

Equation 8

where: Ps = Fraction of herbicide in water Kd= Soil partition coefficient (L/kg) ms = mass of soil (kg dry weight) V = water volume (L) The estimate of the mass of soil is given in Section 4.4.6. The manner in which the channel or drain was refilled could vary between situations. At one extreme the channel could be slowly filled and stand for a period following refilling. At another extreme there could be rapidly flowing water from the time it is refilled and for some time following. These two cases will be considered; 1. In the first case, the total amount of pesticide (the residual in the soil and the amount washed off from the plant) would come to equilibrium between the amount in the soil and the amount in the water. 2. In the second case, the constant replacing of water would lead to a reduction of the amount of pesticide present in a section of channel. The reduction in the amount present would reduce the risk presented in this case. We have therefore considered the first case where the total amount of herbicide in a length of channel is equilibrium between the soil and the channel water as this gives a near worst case scenario. The release of pesticides from soil can be much slower than their absorption – this phenomenon is referred to as a hysteresis effect. Glyphosate binds strongly to soil so it is unlikely the glyphosate that is bound on soil would equilibrate rapidly. This strong binding would further decrease the expected concentration of glyphosate in the channel water, so the predicted concentrations given in this report are likely to be #2482314 File: 2006/1480/1

34

over estimates. Note that in Equation 8, both Kd and ms are on the bottom line. Good data are required for both these parameters to obtain reliable predictions of the PEC.

4.5. Tier 2 calculations for acrolein 4.5.1. Dispersion A chemical injected at a point in a flowing channel is immediately subjected to the processes of turbulent diffusion and dispersion and its concentration tends to become uniform in cross-section. The actual spread or rate of dilution depends on the channel geography and the large-scale turbulence structure of the flow. The distance required before there is complete mixing is typically of the order of hundreds of times the channel width. It is therefore difficult to predict the length of this initial phase of dilution. During the dispersion phase, the chemical behaves as a onedimensional slug of material in the channel; the only significant concentration gradient is that in the direction of flow. The formula below predicts the tracer behaviour (O’Loughlin and Bowmer 1975).

c c  2c U  D 2  Kc t x x

Equation 9

where: D = longitudinal dispersion coefficient; K = the first order rate constant in water (tracer decay); U = velocity of water; and t= number of days since application. Their analytical solution of the formula for case of acrolein (which is non-conservative material) and injected instantaneously gives:

C ( x, t )  Kx x  U (t   )(1  H ) x  Ut (1  H )  0.5 exp( ){erfc[ ]  erfc[ ]} Equation 10 C0 U 4 D(t   ) 4 D(t   ) Where: C(x, t) = concentration of herbicide x m downstream after t hours; C0 = initial applied rate (mg/L); and H = 2KD/U2 There are some unknown parameters in the given formula such as K and D, but estimates can be found for those parameters. The longitudinal dispersion coefficient D depends on the channel depth and the velocity of water and can be founded using the formula of Bowmer (1987): D = 5.9 × U × d, where U is the velocity of the water and d the channel depth.

#2482314 File: 2006/1480/1

35

An estimate of the decomposition rate K can be derived from half-life data. Using the above equations and approximations, estimates of PEC can be calculated in Excel© by making use of the inbuilt normal integral functions.

4.6. Estimation of Drinking Water Equivalent Level (DWEL) Usually there are no data that give estimates of NOAELs for pesticides that are dissolved in drinking water, so an alternative approach is required. The drinking water equivalent level (DWEL) is the concentration of a contaminant in water (in mg/L) for which no adverse (non-carcinogenic health effects) are anticipated if a person is exposed to that concentration over a lifetime. The formula for calculating the DWEL is: DWEL mg/L = Dose (mg/kg/day) × Body Weight (kg) Drinking Water Consumption (L/day)

Equation 11

This represents a best available estimate but this estimate is subject to considerable uncertainty. It is usual practice to include a safety factor of 10 or even 100 is applied to DWELs before making recommendations as to whether the water is potable. The NOAEL values, which are expressed as mg/kg/day, should be converted to mg/L (i.e. ppm) drinking water consumption.

4.7. Estimation of Hazard Quotients Since we have to find out the safety exposure level, the NOAEL values are collected for the most sensitive receptors in each class where data are available. Also, in case of algae where there are limited data available, we work with the lowest available NOAEL value for algae. The HQ is then estimated using the following equation. HQ = PEC/NOAEC or PEC/NOAEL

Equation 12

This was done for all the receptors listed in Table 3 where toxicity data were available and for each of the six herbicides. The PEC was first taken from the Tier 1 estimate and then from the Tier 2 estimate. At this stage, no safety factor was included so the HQs will be less biased. This gives a more realistic estimate of the risk, and it is then the role of management to incorporate safety factors. This contrasts with an aim where the risk assessment is used to assess a safe level of concentration; in that case it would be appropriate to include a safety factor. Where the HQ exceeded 1, there was a significant risk posed by the herbicide. Those cases were marked in red in the results tables. If the HQ was less than 0.1, the risk was considered small, and those cases have been indicated in green. Where the HQ was in the range of 0.1 to 1.0, it was considered that there was a potential #2482314 File: 2006/1480/1

36

risk (assuming a safety factor of 10) so those cases have been highlighted in yellow. To clarify further, while the safety factor is not built in calculations of PEC, we decided that for the interpretation of data it would be useful to colour code the results in such a way that the user is alerted to the consideration of safety factor. Therefore, the decision to colour cells with HQ 0.1 to 1 as green in essence provides the opportunity to the decision maker to consider the case of a safety factor of 10.

#2482314 File: 2006/1480/1

37

5. Results 5.1. Estimation of Tier (excluding acrolein)

1

concentrations

in

channels

The estimation of the PECs for herbicides sprayed onto drains is shown in Table 16. For consistency, the smallest channel has been included in the calculations although typically imazapyr would be used on larger channels as this represents the near worst possible case. Note that the herbicide properties do not affect the Tier 1 PECs. Table 16. PEC estimations in Tier 1 for channel for glyphosate, 2,4-D, imazaypyr and dalapon (1 m bank sprayed on either side of channel) Herbicide

Rate

a.i.

Width

Depth

Runoff

Amount sprayed

Kg in water

Water (ML)

PEC (kg/ML) = mg/L

Glyphosate

9

0.36

2

0.4

0.1

3.2

1.8

2

0.89

Glyphosate

20

0.36

2

0.4

0.1

7.2

4.0

2

1.98

Glyphosate

40

0.36

2

0.4

0.1

14.4

7.9

2

3.96

Glyphosate

40

0.36

8

1.5

0.1

14.4

11.8

12

0.98

2,4-D

10

0.63

2

0.4

0.1

6.3

3.4

2

1.72

Imazapyr

5

0.15

2

0.4

0.1

0.8

0.4

2

0.21

Dalapon (boom)

12

0.74

2

0.4

0.1

8.9

4.9

2

2.44

Dalapon (spot)

10

0.74

2

0.4

0.1

7.4

4.1

2

2.04

For acrolein, the Tier 1 assessment the PEC number would be 2.58 mg/L for the higher application rate and 0.258 mg/L for the lower rate. The higher rate is used in the Murray Valley Area and the lower rate (but with 48 hour exposure) is used in the Torrumbarry Area.

5.1.1.

Tier 1 estimation of concentrations in drains

Only glyphosate and amitrole are used to clear weeds from drains. It is assumed that a depth of 0.1 m would be required before water could be extracted from the drains by a pump. The PECs for the drains are shown in Table 17. Table 17. PEC estimations in Tier 1 for drains for glyphosate and amitole. Calculations consider 1 m bank is sprayed on either side of the drain Herbicide

Rate

a.i.

Width

Depth

Runoff

Amount sprayed

Amount in water

Volume of water (ML)

PEC (kg/ML)

Glyphosate

15

0.36

2

0.1

0.1

5.4

3.0

0.5

5.9

27

0.36

2

0.1

0.1

9.7

5.3

0.5

10.7

11

0.25

2

0.1

0.1

2.8

1.5

0.5

3.0

Amitrole

#2482314 File: 2006/1480/1

38

5.2. Estimation of Tier 1 hazard quotients 5.2.1. Tier 1 hazard quotients for channels (excluding acrolein) The Tier 1 hazard ratios for channels for glyphosate, 2,4-D, dalapon and imazapyr are shown in Table 18 to Table 21. The HQs have been colour codes such that green indicates an HQ 1.0 red (at risk). Glyphosate Table 18 indicates that glyphosate at the PEC level 3.96 mg/L may pose a risk to tomato and other crops and some aquatic life. The HQ for drinking water, tomatoes, Daphnia and general aquatics all exceeded 1, so further investigation (Tier 2) of those scenarios was required. Table 18. Estimation of Tier 1 Hazard Quotient for glyphosate applied to channels. The PEC was 3.96 mg/L

#2482314 File: 2006/1480/1

Receptor

Criterion

Tolerance

HQ

Green Algae (Scenedesmus acutus)

EC50

4.0

0.99

Water flea (Daphnia magna)

Population reduction

1

3.96

Rainbow trout

LC50

8.0

0.50

Aquatic life in fresh water

95% protection

1.2

3.30

Mallard duck

NOAEL

4000

0.00

Rat

NOAEL

300

0.01

Plant and crop

NOAEL

2.2

1.80

Crop (Tomato)

5% reduction

2.7

1.47

Irrigation water

Guideline value

0.1

39.6

Drinking water

Max. permissible conc.

0.7

5.66

39

2,4-D amine The Tier 1 HQs for 2,4-D in channels are shown Table 19. The HQs exceeded 1.0 for some plants and was in the range 0.1 – 1.0 for the others. There was also potential risk posed to some animal species, including rainbow trout. 2,4-D amine at the applied rate is unlikely to cause any problem for mammals (rat) and for aquatic invertebrates, eve in the worst case assessment. However, the HQ exceeded 1.0 for aquatic life, fish, drinking water and crops such as tomato. Further assessments of these scenarios were undertaken in Tier 2 (next stage).

Table 19. Estimation of Tier 1 Hazard Quotient for 2,4-D amine applied to channels. The PEC was 1.72 mg/L

#2482314 File: 2006/1480/1

Receptor

Criterion

Tolerance

HQ

Alga (Selenastrum capricornutum)

NOAEC

26.4

0.07

Duckweed (Lemna gibba)

NOAEC

2.029

0.85

Aquatic Invertebrates

NOAEC

19.7

0.09

Fish (Rainbow trout)

NOAEC

0.0164

104.88

Aquatic life in fresh water

95% protection

0.28

6.14

Rat

NOAEL

3.41

0.50

Crop (Tomato)

No damage

0.15

11.47

Crop (Soybeans)

Damage LOEL

0.22

7.82

Irrigation water

Guideline value

0.03

57.33

Drinking water

Max. permissible conc.

0.07

24.57

40

Dalapon The Tier 1 HQs for dalapon in channels are shown Table 20. The HQs exceeded 1.0 for some plants (corn) and was in the range 0.1 – 1.0 for the others. There was in general low risk to the animal species. Duckweed, shrimp and corn were assessed to carry a high risk and similarly the risk rating for irrigation and drinking water was also high.

Table 20. Estimation of Tier 1 Hazard Quotient for dalapon applied to channels. The PEC was 2.44 mg/L.

#2482314 File: 2006/1480/1

Receptor

Criterion

Tolerance

HQ

Blue-Green algae (Anabaena sp.)

Mortality

10

0.24

Duckweed (Lemna minor)

Physiology general

1.43

1.71

Brown shrimp

EC50

1

2.44

Crustaceans

NOAEL

200

0.01

Gold fish

No death

100

0.02

Bluegills

LC50

105

0.02

Chicken

LD50

500

0.00

Rat

NOAEL

37.5

0.07

Dog

NOAEL

1000

0.00

Crop (Beets)

Injury threshold

7

0.35

Crop (Corn)

Injury threshold

0.35

6.97

Crop (Grapes)

NOAEC

5

0.49

Irrigation water

Guideline value

0.004

610.00

Drinking water

Max. permissible conc.

0.2

12.20

41

Imazapyr The Tier 1 results for channels for imazapyr (Table 21) show that it there is little risk for animals, and rather mixed results for plant species. The alga and duckweed are assessed as being at risk, as was corn. There was also potential risk to soybean. Therefore the greatest risk associated with this herbicide is for algae in the aquatic ecosystems or for irrigation water

Table 21. Estimation of Tier 1 Hazard Quotient for imazapyr applied to channels. The PEC was 0.21 mg/L

#2482314 File: 2006/1480/1

Receptor

Endpoint

Tolerance

HQ

Green algae (Chlorella emersonii)

EC25 Growth

0.02

10.50

Duckweed (Lemna gibba)

EC25 Growth

0.0132

15.91

Water flea (Daphnia sp.)

NOAEC

40.7

0.01

Pink shrimp

NOAEC

132

0.00

Benthic macro invertebrate

NOAEC

18.4

0.01

Rainbow trout, bluegill sunfish

LC50

100

0.00

Northern bobwhite quail

NOAEL

200

0.00

Rat

NOAEL

10,000

0.00

Dog

NOAEL

10,000

0.00

Rabbit

NOAEL

400

0.00

Crop (Corn)

EC15

0.18

1.17

Crop (Soybean)

EC15

0.9

0.23

42

5.2.2. Tier 1 hazard quotients for channels for acrolein A summary of the Tier 1 assessment for acrolein applied to channels (Table 22) shows that the application of acrolein at the higher concentration could exceed all the acceptable values of HQ for irrigation, drinking water, aquatic life and crop plants except Soybean. In the second method (0.3 mg/L) most of the receptors are still at risk. There was therefore a need to undertake further assessment in Tier 2 for such receptors as irrigation value, algae (as represented by Selenastrum capricorutum), aquatic plants (as represented by the duckweed Lemna gibba), fish and drinking water. Table 22. Tier 1 assessment of the risk posed by acrolein applied at two different methods. Higher doses in shorter time and vice versa (red colour in cells denote HQ > 1, amber HQ 0.1 - 1.0 and green HQ < 0.1).

Receptors

Toxicity value (mg/L) NOAEC or LC50 or EC50

High dose-short injection time

Low dose-long injection time

(Application rate of 3 mg/L for 6 hours)

(Application rate of 0.3 mg/L for 48 hours)

HQ1=2.58/Toxicity value

HQ2= 0.258/ Toxicity value

52000

5200

37

3.7

11

1.17

Algae (Selenastrum capricorutum)

0.00005

Duckweed (Lemna gibba)

0.07

Water flea (Daphnia sp.)

0.022

Goldfish

0.0114

230

23

Aquatic life

0.001

2600

260

Rabbit

0.05

52

5.2

Dog

1

2.6

0.26

Crop (Soybean)

15

0.17

0.017

Crops and pasture

1.5 1.72

0.172

8.08

0.808

Drinking water

#2482314 File: 2006/1480/1

0.32

43

5.2.3. Tier 1 hazard quotients for drains Glyphosate The Tier 1 assessment showed that even in the worst case, the use of glyphosate in drains is not likely to cause any problem for mammals (as represented by rat) or for birds. However the HQ exceeded 1.0 for plant crops and tomatoes. The water also has the potential for causing harm to fresh water organisms such as algae and Daphnia, as well as for other purposes such as irrigation and drinking water. The cases where the HQ exceeded 1.0 are considered in the Tier 2 phase of the study. Table 23. Tier 1 assessment of the risk posed by glyphosate to drains. PEC was 10.7 mg/L

#2482314 File: 2006/1480/1

Receptor

Criterion

Tolerance

HQ

Green Algae (Scenedesmus acutus)

EC50

4.0

2.67

Water flea (Daphnia magna)

Population reduction

1

10.7

Fish (Rainbow trout)

LC50

8.0

1.34

Aquatic life in fresh water

95% protection

1.2

8.90

Mallard duck

NOAEL

4000

0.003

Rat

NOAEL

300

0.036

Crop (Tomato)

5% reduction

2.7

3.96

Plant and crop

NOAEL

2.2

4.86

Irrigation Water

Guideline Value

0.1

Drinking water

Max. permissible conc.

0.7

107

15.3

44

Amitrole The tier 1 assessment showed that even under the near worst conditions when amitrole is applied to drains it is not expected to cause harm to mammals (rat), birds and general invertebrates (Table 24). HQs exceeding 1.0 were found for other receptors such as Daphnia, fish, drinking water and health advisory levels. The receptors where the HQ exceeded 1.0 were further assessed in Tier 2. Table 24. Tier 1 assessment of the risk posed by amitrole applied to drains. PEC was 3.0 (red colour in cells denote HQ > 1, amber HQ 0.1 - 1.0 and green HQ < 0.1) Receptor

Criterion

Tolerance

HQ

Green algae (Pseudokirchneriella subcapitat)

EC50

1

3.000

Aquatic Plant

EC50

2.5

1.200

Water flea (Daphnia magna)

Reproduction

0.2

15.000

Fish (Rainbow trout)

NOAEC

100

0.030

Fish

Reproduction

0.2

15.000

Honeybee

NOAEL

100

0.030

Quail, duck

Reproduction

100

0.030

Rat

NOAEL

10

0.300

Dog

NOAEL

62.5

0.048

Irrigation water

Guideline vlaue

0.002

1500.000

Drinking water

Guideline value

0.001

3000.000

5.3. Estimation of Tier 2 concentrations 5.3.1. Tier 2 Estimation of concentrations in channels (excluding acrolein) The concentration of glyphosate was estimated following the process outlined in Section 4. As discussed in that section, no data are available on the fraction of pesticide that is washed off the plant when the water returns to the channel. A typical calculation of the PEC is given in Table 25, where it is assumed that there is a 60% wash-off. A similar calculation was also performed where the wash-off fraction was set at 20% or 1%. Note that in the calculations, the fraction of spray that was intercepted by water is assumed to equilibrate with the ‘schlick’ at the bottom of the channel, so in the case of glyphosate (with a high Koc), most of that component would be bound by the soil. This would not be so for pesticides like 2,4-D with a low Koc.

#2482314 File: 2006/1480/1

45

Table 25 Estimation of concentration of glyphosate in channel water. Following an application of 40 L/ha with 0.36 a.i. with 60% interception by plants, and assuming 60% of the residual herbicide is washed off in channel 2 m wide and 0.4 m deep Water

Plant

soil

Total

+2% run-off Interception

40%

60%

0%

100%

Amount applied (kg/ha)

6.05

8.64

0

14.69

Days after exposed

4

4

4

Half-life (days)

70

2.5

47

Faction remaining

0.96

0.33

0.94

Amount remaining (kg/ha)

5.81

2.85

0

Wash-off

1

0.6

0

Fraction in water column after sorption to soil layer at channel bottom

0.003

1

0

Amount released to water (kg/ha)

0.01743

1.71

0

1.72743

Water volume (ML/ha sprayed area)

4

4

4

4

Final concentration (mg/L)

-

-

-

0.43

In the case of imazypyr and dalapon, it is assumed that the channel is mainly covered by arrowhead and water level is very low (typically 1, amber HQ 0.1 – 1.0 and green HQ < 1). * indicates values based on EC50. 3 km 1.05

Distance from the injection point (km) 5 km 7 km 10 km 15 km 20 km 0.58 0.32 0.13 0.029 0.006

25 km 0.0014

30 km 0.000072

35 km 0.000016

Safe distance# 40 km 0.000003

38400

21120

11600

6400

2600

600

130

28

6.6

1.5

0.33

0.001 0.0114 0.07 0.05 0.32 1

1920 168 27 38 6.00 1.92

1056 93 15 21 3.30 1.06

580 51 8.3 11.6 1.81 0.58

320 28 4.6 6.4 1.00 0.32

130 11 1.9 2.6 0.41 0.13

30 3 0.43 0.60 0.09 0.03

6.5 0.57 0.09 0.13 0.02 0.01

1.4 0.12 0.02 0.03 0.00 0.00

0.33 0.03 0.00 0.01 0.00 0.00

0.01 0.00 0.00 0.00 0.00 0.00

0.00 0.00 0.00 0.00 0.00 0.00

Crops and pasture

1.5

1.28

0.70

0.39

0.21

0.09

0.02

0.00

0.00

0.00

0.00

0.00

Crop (Soybean)

15

0.13

0.07

0.04

0.02

0.01

0.00

0.00

0.00

0.00

0.00

0.00

Toxicity mg/L

HQ 1 km 1.91

Algae (Selenastrum capricorutum*)

0.00005

Aquatic life Goldfish Duckweed (Lemna gibba*) Rabbit Drinking water Dog

Species PEC (mg/L)

#These “Safe Distances” are only relevant for the flow velocity used here (463m/h) and would markedly change with change in flow conditions.

#2482314 File: 2006/1480/1

56

Table 34. Hazard Quotient for 9 receptors at different distances from source of application of 0.3 mg/L of acrolein. (red colour in cells denote HQ > 1, amber HQ 0.1 – 1.0 and green HQ < 1). * indicates values based on EC50. Toxicity Species

HQ

Distance from the injection point (km)

PEC (mg/L) Algae (Selenastrum capricorutum*) Aquatic life Goldfish Rabbit Duckweed (Lemna gibba*) Drinking water Dog

0.00005 0.001 0.0114 0.05 0.07 0.32 1

0.116 2320 116 10.2 2.32 1.66 0.36 0.12

0.023 460 23 2.02 0.46 0.33 0.07 0.02

0.0048 80 4.0 0.35 0.08 0.06 0.01 0.00

0.0021 42 2.1 0.18 0.04 0.03 0.01 0.00

0.00009 1.8 0.09 0.01 0.00 0.00 0.00 0.00

Safe distance 12 km# 0.000018 0.04 0.00 0.00 0.00 0.00 0.00 0.00

Crops and pasture

1.5

0.08

0.02

0.00

0.00

0.00

0.00

Soybean

15

0.01

0.00

0.00

0.00

0.00

0.00

mg/L

1 km

3 km

5 km

6 km

10 km

#These “Safe Distances” are only relevant for the flow velocity used here (463m/h) and would markedly change with change in flow conditions.

#2482314 File: 2006/1480/1

57

5.4.3. Tier 2 hazard quotients for drains Glyphosate The hazard quotients for receptors to glyphosate applied in drains are shown in Table 35Table 35. At the highest wash-off rate of 60% there is the potential of harm to Daphnia and to aquatic ecosystems. The concentration of glyphosate in the water would exceed guidelines for both irrigation water and drinking water. Even at 10% run-off, the concentration still exceeds the guideline for irrigation water despite the HQ for tomato being only 0.072 and that for irrigation water being 1.944. This would imply that the guideline value of 0.1 mg/L for irrigation water is possibly too low and may need review. Table 35. Tier 2 hazard quotients for glyphosate applied to drains. The wash-off factors of 60%, 10% and 5% correspond to PECs of 1.55, 0.26 and 0.032 mg/L. Also shown is the effect of 10-fold and 100-fold dilution of the highest wash-off factor Receptor

Criterion

Tolerance

Washoff = 60%

Washoff = 10%

Washoff = 5%

Dilution 10-fold

Dilution 100-fold

Algae

EC50

4.85

0.320

0.054

0.007

0.032

0.003

Water flea (Daphnia magna)

Population reduction

1

1.550

0.260

0.032

0.155

0.016

Fish (Rainbow trout)

LC50

8.2

0.189

0.032

0.004

0.019

0.002

Aquatic life in fresh water

95% protection

1.2

1.292

0.217

0.027

0.129

0.013

Mallard duck

NOAEL

4000

0.000

0.000

0.000

0.000

0.000

Rat

NOAEL

300

0.005

0.001

0.000

0.001

0.000

Plant and crop

NOAEL

2.2

0.705

0.118

0.015

0.070

0.007

Crop (Tomato)

5% reduction

2.7

0.574

0.096

0.012

0.057

0.006

Irrigation water

Guideline value

0.1

15.500

2.600

0.320

1.550

0.155

Drinking water

Max. permissible conc.

0.7

2.214

0.371

0.046

0.221

0.022

#2482314 File: 2006/1480/1

58

Amitrole Table 36 shows that the HQs for the use of amitrole in drains exceed 1.0 for Daphnia, and grossly exceed the value for irrigation water and drinking water. It may also cause harm to the green alga, noting that that HQs were based on EC50s and not NOAECs. Table 36. Tier 2 hazard quotients for amitrole applied to drains. The wash-off factors of 95%, 20% and 1% correspond to PECs of 1.52, 0.57 and 0.38 mg/L. Also shown is the effect of 10-fold and 100-fold dilution of the highest wash-off factor Receptor Criterion Tolerance Wash- WashWash-off Dilution Dilution off = off = = 1% 10-fold 100-fold 95% 20% Green algae (Pseudokirchneriella subcapitat)

EC50

1

1.52

0.57

0.38

0.152

0.057

Aquatic Plant

EC50

2.5

0.608

0.228

0.152

0.0608

0.0228

Water flea (Daphnia magna)

Reproduction

0.2

7.6

2.85

1.9

0.76

0.285

Fish (Rainbow trout)

NOAEC

100

0.0152

0.0057

0.0038

0.00152

0.00057

Fish

Reproduction

0.2

7.6

2.85

1.9

0.76

0.285

Honeybee

NOAEL

100

0.0152

0.0057

0.0038

0.00152

0.00057

Quail, duck

Reproduction

100

0.0152

0.0057

0.0038

0.00152

0.00057

Rat

NOAEL

10

0.152

0.057

0.038

0.0152

0.0057

Dog

NOAEL

62.5

0.02432

0.00912

0.00608

0.002432

0.000912

Irrigation water

Guideline value

0.002

760

285

190

76

28.5

Drinking water

Guideline value

0.001

1520

570

380

152

57

#2482314 File: 2006/1480/1

59

6. Discussion 6.1. Comparison of predictions with monitoring data There has been some monitoring in channels and drains reported in a Goulburn River Audit (2005). A summary of the results are shown in Table 37. The glyphosate levels observed in the drains were consistent with those modelled in this study apart from one extreme value of 10.5. This high value was found following sampling almost immediately after spraying and there may have been incomplete mixing including absorption by the soil at the bottom of the drain. The predicted amitrole concentrations are in the same range of the monitoring data, especially taking into account the modelled predictions were based on several assumptions. These similarities between the predicted and monitored concentrations provide confidence in the modelling methods used in this study.

Table 37 Range of monitored data and predicted values for glyphosate, 2,4-D and amitrole Reference

Treatment

Channel or Drain

Glyphosate

2,4-D

Amitrole

Table 1

Glyphosate and amitrole

Drain

< LOD - 0.08

NT

< LOD – 0.029

Table 2

Glyphosate

Drain

0.01 – 0.94

NT

0.002 – 1.06

Table 3

Glyphosate

Drain

< LOD – 10.5

NT

NT

Table 4

2,4-D

Channel

NT

< LOD

NT

Table 5

2,4-D

Channel

NT

< LOD – 2.96

NT

Table 6

2,4-D

Channel

NT

< LOD

NT

This study

Glyphosate and amitrole

Drain

0.018 – 1.54

NA

0.29 – 1.42

This study

Glyphosate and 2,4-D

Channel

0.002– 0.43

0.51 – 0.81

NA

There were no monitoring data for dalapon or imazapyr.

6.2. Comparison of herbicide effects 6.2.1. Glyphosate Glyphosate is widely used in the control of weeds from waterways. It is used in a variety of ways including spraying when the water levels are low or in target spraying with a hand gun to remove potentially troublesome patches of weeds. The only threat posed by glyphosate was to irrigation water, and that threat occurred only when a high wash-off value was assumed. There was a similar result for drains, where again the main threat posed was to irrigation water. However, if a 60% wash-off is assumed (in the absence of actual #2482314 File: 2006/1480/1

60

data) potential threat to Daphnia, aquatic ecosystems and drinking water was identified. The same wash-off factor was used for glyphosate sprayed whether it was sprayed in channels or in drains. This is only an approximation as plant parts above the water level in the drains may be sprayed but would be out of the water for several days because glyphosate is slow acting. The levels of water in channel or drain at the time of a given assessment would determine the actual wash-off factor and should be considered prior to decision making. Further risk reduction can be achieved by using a hand gun to target patches of potentially troublesome weeds. Such a reduction in glyphosate application has the added beneficial effect of reducing the amount of carrier that is being applied.

6.2.2. 2,4-D amine 2,4-D has the potential for causing more harm than glyphosate. It is more mobile and less sensitive to the wash-off fraction. Channel water treated with 2,4-D, if undiluted, has the potential to affect crops and aquatic life for all wash-off fractions. Similarly, there is some potential for risk to fish even after 10 time dilution. The risk for drinking water also exists even after 10 times dilution of channel water under the scenario conditions used in the report. A longer withholding time before water is released back in channel should be considered for this herbicide.

6.2.3. Amitrole Amitrole when applied to drains has the potential to cause harm to Daphnia when it is applied at a high rate and if the amount sprayed on plant washes off the plant surfaces in waterways. There is no data available on this process. Caution should be used in allowing drainage water from a drain that has been sprayed with amitrole should not be allowed to enter directly into a wetland. There is also the potential that it will exceed the recommended levels for irrigation water. Irrigation from drainage water following spraying with amitrole should therefore be undertaken with caution.

6.2.4. Dalapon Due to relatively long half-life in water (30 days) and very low Koc (1 g/mL), dalapon may readily enters into water from soil and thus plays an important role in determining PEC number. Moreover, dalapon may also be released from foliage where it has a half-life of 37 days in plant tissue. There is some uncertainty as to its half-life in dead plants. Corn is sensitive to dalapon and could be at risk. Even with a very low wash-off fraction the HQ still exceeds 0.1. Grapes and beet are less sensitive, but if there is a high wash-off fraction the HQ still exceeds 0.1. It also could possibly represent a threat to invertebrates (as represented by brown shrimp) as the HQ exceeds 0.1.

#2482314 File: 2006/1480/1

61

6.2.5. Imazapyr Since the half-life of imazapyr in water is short (4 days), it’s concentration in a channel will reduce quite rapidly with time. Any potential adverse effects from imazapyr could be alleviated by holding water in the channel for a period of time – for example by retaining water in the channel for 8 days would reduce the PEC by a factor of 4. By using such a strategy imazapyr could be used safely for the removal of weeds in a dry channel. The strategy for reducing the risk by retaining the water in the channel depends on the accuracy of the half-life estimation of imazapyr in water. It is therefore recommend that some monitoring be undertaken to validate the PEC values suggested in this report and also to confirm the effective half-life of imazapyr in channel water.

6.2.6. Acrolein Acrolein is a potent herbicide and has the potential to cause harm many kilometres from the injection point. If applied at 3.0 mg/L level, for general aquatic life this could be 25 km and for algae as much as 35 km. The position is much safer if acrolein is injected at the lower rate of 0.3 mg/L. Were acrolein to be applied to drains, there is the potential for longer half-life due to the lack of turbulence as well as the threat to its receiving water body. Acrolein should not be used in drains – we understand this is G-MW policy.

6.2.7. Overview In general there is only small number of cases where there is potential for harm to ecosystem from the herbicides. The greatest threat is to exceeding the irrigation water guidelines. The results of this assessment are very much subject to the input values, which highlights the need for better data on critical factors such as the wash-off fraction form plant surface and half-life of dissipation from water (e.g. for imazapyr).

6.3. Extrapolation to other receptors A comparison of tolerance values across receptors is presented in Table 38 to provide a snapshot of relative tolerance of different herbicides for different receptors. A given herbicide may be more toxic for one receptor and not the other. This makes generalised extrapolation to other species difficult. Although this report considers a limited range of receptors, a method being developed by CSIRO (Morton et al. 2007) would enable the extrapolation of end points from one species to another. The ability to extrapolate would enable much more general statements to be made concerning the findings given in this report.

#2482314 File: 2006/1480/1

62

Table 38. Comparison of tolerance values across receptors

Alga

Glyphosate

2,4-D

Dalapon

Imazapyr

Amitrole

Acrolein

4.85 (T)

26.4 (H1)

10

0.02

1(W1)

0.00005 (T)

2.029 (H1)

1.43

0.0132

2.5(W2)

0.07 (T)

Lemna Daphnia

1

19.7

1

40.7

1

0.022

Rainbow trout

8.2

0.0164

100

>100

100

0.0114

Crops

2.2 (H2)

0.22 (Q)

0.35

0.1

Tomatoes

2.7 (S)

0.15 (F)

Irrigation water

0.1 (G)

0.03 (G)

1.5 (G)

0.004

0.002 (G)

N/A

F= Fagliari et al. (2005) G= Goulburn Murray manual, H1= Hughes et al. (1990), H2= Hutson and Roberts (1987), Q= Que et al. (1981), S= Santos and Gilreath (2006), T= Tomlin (2000), W1= Wang et al. (1990), W2 = Wolf (2001). A comparison of the risks posed by each herbicide (except acrolein) is shown in Table 39. Each herbicide has its benefits and problems. The problems posed in drains are harsher due to the less dilution, and the potential of the water to reach natural carriers. This can be seen by a comparison of the risks posed by glyphosate in channels and drains. In general, there are relatively small number of scenarios where potentially risks is posed. Glyphosate and amitrole pose little threat to the environment, but there is potential of harm to corn. Imazapyr poses very little risk to animals, and its short half-life in water could potentially be used to advantage. Amitrole applied to drains could potentially harm Daphnia and fish unless it is diluted. Table 39. Comparison of near worst risks posed by of herbicides to key receptors when applied at maximum rates Species

Green algae (Chlorella emersonii)

Channel

Drain

Glyphosate

2,4D

Dalapon

Imazapyr

Glyphosate

Amitrole

0.089

0.045

0.089

1.55

0.32

1.52

0.586

0.622

2.348

Duckweed (Lemna gibba) Freshwater flea (Daphnia)

0.43

0.06

0.004

0.001

1.55

7.60

Rainbow trout, bluegill sunfish

0.052

72.561

0.009

0

0.189

0.015

Rat

0.001

0.349

0.024

0

0.005

0.152

6.4. Other risk not considered in this report This report has considered only the active herbicide component of the products that are applied. Typically the products contain some type of carrier and possibly a surfactant (adjuvants), with a result that the fraction of active ingredient may be #2482314 File: 2006/1480/1

63

significantly less than 1. Furthermore there is no consideration of toxicity of mixture of toxicants in the present assessment. There is research that shows that the product can be more toxic to amphibians than the so -called active ingredient. This is particularly so for Roundup® as compared to glyphosate (Schmuck 1994). This is highlighted by the data shown in Table 40. This problem has been recognised by G-MW and as a result they have changed their formulation that contains glyphosate (pers. comm. workshop in Tatura November 2007). Table 40. Toxicity of components of Roundup® to rainbow trout (after Bowmer et al. 1998)

Chemical

24 h LC50

Surfactant

2.1 mg/L

Glyphosate

140 mg/L

Roundup®

8.3 mg/L

6.4.1. Acute versus chronic risks This study has focussed on acute risks which is appropriate for the assessment of local effects. There may also be a much lower threshold for chronic risks. For example, the acute NOAEL for rainbow trout exposed to amitrole is 100 mg/L (Wolf 2001), but the limit for reproduction is only 0.2 mg/L. Caution must therefore be applied not only to the acute levels but also to the chronic levels that could arise as the sum of many small effects upstream of the receiving water.

6.4.2. Daughter products It is well known that pesticides form breakdown products, some of which can be more or less toxic and persistent. Scribner et al. (2003) reported on 154 samples that were collected from 51 streams in nine Midwestern States during three periods of run-off. Results showed that glyphosate was detected in 55 (36 percent) of the samples, and aminomethylphosphonic acid (a degradation product of glyphosate) was detected in 107 (69%) of the samples. There is evidence that daughter products do persist and their presences should not be ignored. The problem of daughter products needs to be considered as part of the overall herbicide fate considerations, and this should be factor in the choice of herbicides.

#2482314 File: 2006/1480/1

64

6.5. Closing remarks The assessment of risks posed by herbicides to non-target receptors involve many factors and requires estimates of many parameters. When assessing the risks, ideally a thorough understanding of the processes should be obtained. However, because of the complexity of the processes involved, an approximation to reality is required in most cases. In choosing a modelling process, there has to be a compromise between a complex which requires a large amount of data and many parameters on the one hand versus a simple model that is incomplete but perhaps more robust. In this report we have sought to have a model that uses only data that are available or are potentially available. There are numerous scenarios that could be considered. For example, drains with a small depth of water might have a much higher concentration of herbicides than the scenarios considered in this report. Another aspect not fully explored in this report is the potential for dilution before the herbicides reach natural carriers and wetlands. Before making any decisions on any change of management practices, the results on risk assessments made in this report (for various receptors in selected specific scenarios at the point of treatment) should be carefully considered. Such an analysis should incorporate dilution factors and other risk moderating or moderating factors such as channel/drain dimensions (especially depth of water), buffer between treated area and receiving environments, flow velocity and the withholding period. There are opportunities for GM-W to refine the risk estimates by addressing some critical data gaps identified in the report and manage the risks by making safer choices based on the risk rating provided in this report.

#2482314 File: 2006/1480/1

65

7. References Abbott PJ (1994) INCHEM Environmental health Criteria 158, Amitrole Available at: http://www.inchem.org/documents/ehc/ehc/ehc158.htm Aly O M and Faust S D (1964) Studies on the fate of 2,4-D and ester derivatives in natural surface waters. J. Agric. Fd. Chem. 12: 541-546 (non vide), referenced by Weed Control Methods Handbook, The Nature Conservancy, Tu et al, section 7a, update, 2,4-D (2001)) Available at: http://tncweeds.ucdavis.edu/products/handbook/10.24D.pdf Anon (2005) Water quality of Yellowstone County Available at: http://agr.state.mt.us/pestfert/groundwater/yellowstone_yellowstone.pdf Accessed May 2007 ANZECC and ARMCANZ (2000) “Australian and New Zealand Guidelines for Fresh and Marine Water Quality” the Guidelines V.1 Chapter 4 Primary industries Available at: http://www.environment.gov.au/water/quality/nwqms/pubs/wqg-ch4.pdf Azzouzi E. et al. (1998) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) Bellingham (2005) Town of Bellingham Department of Public Works Drinking Water Quality Consumer Information Report available at: http://www.bellinghamma.org/DPW/CCR%202005.pdf Accessed May 2007. Bowmer KH and Sainty GR (1991) The acrolein review use safety and environmental implication NSW department water resources in CSIRO Division of Water Resources, Griffith, 55 pp Bowmer KH, Korth W, Scott A, McCorkelle G, Thomas M. (1998) Pesticide monitoring in the irrigation areas of south-western NSW 1990-1995. CSIRO Land and Water Bowmer, K. (1987). Herbicides in surface water. In: Hutson, D H; and Roberts, T R (Eds.). Herbicides, (Ch.9): 271-355 311 refs, illus. Bridie et al. 1979 ( non vide, referenced by Geoff Siemering, Nicole David, Jennifer Hayworth and Amy Fran z, 2005, San Francisco Estuary Institute, Aquatic Pesticides Monitoring Program) Available: http://www.sfei.org/apmp/reports/71_APMP_litreview_FINAL.pdf Briggs G.G. Bromilow R.H. and Evans A.A. (1982) (non vide, referenced by GLEAMS) Available at:http://www.tifton.uga.edu/sewrl/Gleams/glm30pst.pdf Cheng H. H. Bailery G. W. Green R. H. Spencer W.F. (1990) Pesticide in the Soil Environment: Processes, Impacts, and Modelling, Number 2 in the Soil Science Society of America Book Series, published by: Soil Science Society of America , Inc. Madison, Wisconsin Cohen Z. (1984) Memorandum from Stuart List of potential groundwater contaminants Office of Pesticides and Toxic Substances. Washington, DC. Photocopy. (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/dalaponext.html#20 Cox C. (1996). Herbicide fact sheet Imazapyr Journal of Pesticide Reform 16(3): 1-5. Cyanamid (1997) (non vide, referenced by Pless P. (2005) use of Imazapyr herbicide to control invasive cordgrass (spartona spp.) in the San Francisco estuary) available at: http://www.spartina.org/referencemtrl/SF-Imazapyr-EA.pdf Dickens and Wehtje (1986) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy)

#2482314 File: 2006/1480/1

66

EXTOXNET (1996) Dalapon. Available at: http://extoxnet.orst.edu/pips/dalapon.htm EXTOXNET (1996) Amitrole Available at: http://extoxnet.orst.edu/pips/amitrole.htm EXTOXNET(1996) 2,4-D Available at: http://extoxnet.orst.edu/pips/24-D.htm EXTOXNET(1996).Glyphosate Available at:http://extoxnet.orst.edu/pips/glyphosa.htm Fagliari JR, de Oliveira RS. Jr and Constantin J. (2005) impact of sublethal doses of 2,4-D, on tomato yield, Environmental Science and Health B. 40(1): 201-206 Availableat: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=A bstractPlus&list_uids=15656182&query_hl=8&itool=pubmed_docsum FAO (1997) Corporate Document Repository, organized by Agriculture Department, 2,4-D and its salt and ester Available at: http://www.fao.org/docrep/W8141E/w8141e0n.htm Fowlkes et al. (2003) (non vide, referenced by Pless P. (2005) use of Imazapyr herbicide to control invasive cordgrass (spartona spp.) in the San Francisco estuary) available at: http://www.spartina.org/referencemtrl/SF-Imazapyr-EA.pdf Goulburn River Audit (2005) Residue testing in response to Goulburn River Audit, 2005. Goulburn-Murray Water Document 2034472 Hartley D. and Kidd H. eds. (1983) The Agrochemicals handbook Nottingham England: Royal Society of Chemistry (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/dalaponext.html#20 Hgonzales (2004) Imazapyr, Fact sheet available at: http://www.spartina.org/referencemtrl/ImazapyrFactSheet.pdf Howard PH. (1989) (Non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/24d-captan/amitrole-ext.html Howard, P.H. (1989) Handbook of Environmental Fate and Exposure Data for Organic Chemicals V3: Pesticides. Lewis Publishers, Chelsea MI. (non vide, referenced by EXTOXNET, 1993) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/dalaponext.html#20 HSDB (2003) (non vide, referenced by USEPA (2003) CAS No.107-02-8, In support of Summary Information on the Integrated Risk Information System (IRSI)) Available at: http://www.epa.gov/iris/toxreviews/0364-tr.pdf Hughes JS (1990) the toxicity of 2,4-D, 2-ethylhexyl ester to Anabaena flos-aquae (non vide, referenced by INCHEM, (1997) pesticide residue in food toxicological and environmental evaluation) Available at: http://www.inchem.org/documents/jmpr/jmpmono/v097pr16.htm Hughes JS, Williams TL and Conder LA (1997) (non vide, referenced by INCHEM (1997) pesticide residue in Food Toxicological and Environmental Evaluation) Available at: http://www.inchem.org/documents/jmpr/jmpmono/v097pr16.htm Hutson DH. and Roberts TR (1987) herbicides-progress in pesticide biochemistry and toxicology; V6, pp.300 INCHEM (1994) IPCS Amitrole health and safety guide No.85 Available at: http://www.inchem.org/documents/hsg/hsg/hsg85_e.htm INCHEM (1997) Pesticide residue in food, toxicological and environmental evaluation Available at: http://www.inchem.org/documents/jmpr/jmpmono/v097pr16.htm Johnson WG. Lavy TL. and Gbur EE. (1995) (non vide, referenced by Weed Control Methods Handbook, The Nature Conservancy, Tu et al, section 7a, update, 2,4-D (2001)) Available at : http://tncweeds.ucdavis.edu/products/handbook/10.24-D.pdf Kookana, R., Barnes, M and Correll, R. and Kibria, G. (2003). First tier assessment of the #2482314 File: 2006/1480/1

67

risks associated with pesticides used in Goulburn-Murray Irrigation areas. Report prepared under a research collaboration agreement between G-MW and CSIRO. 103p http://www.g-mwater.com.au/downloads/R_D/pesticides_riskassessment.pdf Landstein et al. (1993) (non vide, referenced by Pless P. (2005) use of Imazapyr herbicide to control invasive cordgrass (spartona spp.) in the San Francisco estuary) available at: http://www.spartina.org/referencemtrl/SF-Imazapyr-EA.pdf Langdon K. Warne M. St. J. Kookana R. Karkainnen M. (2006) Selection of organic chemicals to be analysed in the “Risk assessment of pathogens and organic contaminants in biosolids” project Layton R (2000) GLEAMS (Groundwater Loading Effects of Agricultural Management Systems) available at: http://www.tifton.uga.edu/sewrl/Gleams/glm30pst.pdf Leonard O. A., Linder L. A., Lang A. H. (1964) Toxicity of several herbicides to grapes rooting applied to the roots and to the shoots. American Journal of oenology and viticulture 15 pp. 206-213 abstract available at http://www.ajevonline.org/cgi/content/abstract/15/4/206 Malik J. Barry G and Minireview KG (1989) (non vide, referenced by EXTOXNET (1996) Glyphosate) Available at: http://extoxnet.orst.edu/pips/glyphosa.htm Mallipudi et al. (1991) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) Mangels (1991) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) Mangels and Ritter (2000) (non vide, referenced by Pless P. (2005) use of Imazapyr herbicide to control invasive cordgrass (spartona spp.) in the San Francisco estuary) available at: http://www.spartina.org/referencemtrl/SF-Imazapyr-EA.pdf McDowell et al. (1997) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) Meister, R.T. (1992) Farm Chemicals Handbook '92. Meister Publishing Company Willoughby OH. (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/dalaponext.html#20 Miller CS. and Hall WC (1961) (non vide, referenced by INCHEM (1974) IPCS, Amitrole) Available at: http://www.inchem.org/documents/jmpr/jmpmono/v074pr03.htm Morton R, Warne M St.J and Correll RL (2007). Taxonomic extrapolation of Ecological toxicants. Environmetrics (in review). Newton M, Howard KM, Kelpsas BR, Danhaus R, Lottman CM, Dubelman S (1984) Fate of glyphosate in an Oregon forest ecosystem. J Agric Food Chem 32: 1144-1151 O’Brien M.C. and Prendeville G.N. (1979) Effect of Herbicides on Cell Membrane Permeability in Lemna minor journal of Weed Resource. 19 (6) p.p. 331-334 (non vide, referenced by Ecotoxicology references) available at: http://www.pesticideinfo.org/List_AquireAll.jsp?Species=507&offset=250 O'Loughlin E. M. and Bowmer K. H. (1975) Dilution and decay of aquatic herbicides in flowing channels. Journal of Hydrology 26: 217-235. Pless P. (2005) use of Imazapyr herbicide to control invasive cordgrass (spartona spp.) in the San Francisco estuary available at: http://www.spartina.org/referencemtrl/SF-Imazapyr-EA.pdf PMEP (1984) Chemical Fact sheet for: Amitrole Herbicide Profile 5/84 Available at: http://pmep.cce.cornell.edu/profiles/herb-growthreg/24-d-butylate/amitrole/herb#2482314 File: 2006/1480/1

68

prof-amitrole.html Que HSS., and Sutherland RG (1981) The Phenoxyalkanoic Herbicides, Volume I: Chemistry, Analysis, and Environmental Pollution. CRC Press, Inc., Boca Raton, Florida. 319 pgs (non vide, referenced by http://tncweeds.ucdavis.edu/products/handbook/10.24-D.pdf Raymond, S., Nugegoda, D. and Kibria, G. (2006).The effects of pulse exposure of six agricultural chemicals (including four herbicides used by G-MW) on the early life stages of selected native fish from the Goulburn-Murray River regions. Report prepared under a research collaboration agreement between G-MW and RMIT University. 89p. (peer reviewed by Dr John Chapman, Editor in Chief, Australasian Journal of Ecotoxicology).Summary of the paper on G-MW web site, http://www.gmwater.com.au/downloads/R_D/Pesticides_in_Irrigation_Water_Workshop_Abstra ct.pdf Renzo (2000) FAO specification and evaluations for plant protection products, Glyphosate, 2000-2001 Available at: http://www.fao.org/ag/AGP/AGPP/Pesticid/Specs/docs/Pdf/new/glypho01.pdf#sear ch=%22Noael%20for%20duck%20glyphosate%22 Rice D. (1997) Public Health Goal for Dalapon in Drinking Water prepared by Environmental Toxicity Section Office of Environmental Health Hazard Assessment California EPA Available at: http://www.oehha.ca.gov/water/phg/pdf/dalap_c.pdf Ritter (1989) (non vide, referenced by Abbott P.J. (1994) INCHEM Environmental health Criteria 158, Amitrole) Available at: http://www.inchem.org/documents/ehc/ehc/ehc158.htm Ross MA and Childs DJ (2007) Herbicide mode of action summary Available at: http://www.ces.purdue.edu/extmedia/WS/WS-23-W.pdf Rueppel ML, Brightwell BB, Schaefer J and Marvel JT (1977) (non vide, referenced by Weed Control Methods Handbook the Nature Conservancy) Available at: http://tncweeds.ucdavis.edu/products/handbook/14.Glyphosate.pdf Saenz, M.E., W.D. Di Marzio, J.L. Alberdi, and M. Carmen Tortorelli, (1997). Effects of Technical Grade and a Commercial Formulation of Glyphosate on Algal Population Growth. Bull.Environ.Contam.Toxicol. 59(4):638-644

Santos BM. Gilreath JP Esmel CE and Siham MN (2006) effects of sublethal glyphosate rates on fresh market tomato Crop Protection 26: 89-91. Schuette J (1998) Environmental fate of Glyphosate Environmental management and Pest Management Department of Pesticide Regulation, Available at: http://www.cdpr.ca.gov/docs/empm/pubs/fatememo/glyphos.pdf Scribner A.E. Battaglin W.A. Dietze J.E. and Thurman E.M. (2003) Reconnaissance data for glyphosate, other selected herbicides, their degradation products, and antibiotics in 51 Streams in nine midwestern States, 2002 U.S. Geological Survey Report 03217. Shaner, D.L. and O’Connor, S.L. (1991). Eds. The Imidazolinone Herbicides. CRC Press Boca Raton, FL. Siemering G., Nicole David, Jennifer Hayworth and Amy Fran z, 2005, San Francisco Estuary Institute, Aquatic Pesticides Monitoring Program) Available: http://www.sfei.org/apmp/reports/71_APMP_litreview_FINAL.pdf Tomlin CDS (2000). The Pesticide Manual 12th Edition (ed. CDS Tomlin, 2002). British Crop Protection Council. Tu M. et al. (2001) Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy. http://tncweeds.ucdavis.edu US DOA. (1984) Department of Agriculture, Forest Service. (August 1984) Herbicides #2482314 File: 2006/1480/1

69

Agricultural handbook no. 633: Pesticide background statements. V1 Prepared by Mitre Corporation, McLean, VA. (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryldicrotophos/dalapon-ext.html#20 USEPA (1992) (non vide), referenced by summary table Available at : http://www.willapabay.org/~coastal/nospartina/impacts/eis1993/table8.pdf USEPA (2000) (none vide) available at: http://www.cdpr.ca.gov/docs/envjust/pilot_proj/lag/meetings/2006_mtgs/EJ_prot_att chmts.pdf USEPA (2003) CAS No.107-02-8, In support of Summary Information on the Integrated Risk Information System (IRSI)) Available at: http://www.epa.gov/iris/toxreviews/0364tr.pdf USEPA (2005) 2,4- D Amine Red fact Available at: http://www.epa.gov/oppsrrd1/REDs/factsheets/24d_fs.htm USEPA (2006) Available at :http://www.epa.gov/IRIS/subst/0057.htm USEPA (2007) Ground water and Drinking water: http://www.epa.gov/safewater/dwh/csoc/glyphosa.html USEPA (June 2005) pesticides, 2,4-D RED fact Available at: http://envirocancer.cornell.edu/turf/pdf/24d_fs.pdf USEPA, 2,4-D Acid (chemical summary) toxicity and exposure assessment for children’s health Available at: http://www.epa.gov/teach/chem_summ/24D_summary.pdf Venkataraman G.S. and Rajyalakshmi B. (1972) Relative Tolerance of Nitrogen-Fixing BlueGreen Algae to Pesticides Indian Journal of Agricultural Science 42 (2) p.p119-121 (non vide, referenced by Ecotoxicology references) available at: http://www.pesticideinfo.org/List_AquireAll.jsp?Species=546&offset=50 Vizantinopoulos and Lolos (1994) (non vide, referenced by Tu M. et al. (2001) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) Vogue PA, Kerle EA and Jenkins JJ (1994) OUS Extension Pesticide Properties Pesticide .Available at: http://npic.orst.edu/ppdmove.htm Wang W, Gorsuch J,W, and Lower W,R, (1990) Plants for Toxicity Assessment, Multispecies Methods of Testing for Toxicity: Use of the Rhizobium-Legume Symbiosis in Nitrogen Fixation and Correlations Between Responses by Algae and Terrestrial Plants ASTM STP 1091, Philadelphia, PA :69-84 Wauchope R,D, Buttler T,M, Hornsby AG, Augustijn-Beckers PWM, and Burt JP (1992) (non vide, referenced by EXTOXNET, 1996) Available at: http://extoxnet.orst.edu/pips/glyphosa.htm. Wauchope R.D. Buttler T.M. Hornsby A.G. Augustijn-Beckers P.W.M. and Burt J.P. (2001) (non vide, referenced by GLEAMS) Available at: http://www.tifton.uga.edu/sewrl/Gleams/glm30pst.pdf Wauchope RD Buttler TM, Hornsby AG, Augustijn-Beckers PWM and Burt JP (2001) (non vide, referenced by GLEAMS) Available at: http://www.tifton.uga.edu/sewrl/Gleams/glm30pst.pdf Weber H and Patschke K (1978) (non vide, referenced by Abbott PJ (1994) INCHEM Environmental health Criteria 158, Amitrole) Available at: http://www.inchem.org/documents/ehc/ehc/ehc158.htm Willis GH, Spencer WF and McDowell LL (1980) (non vide, referenced by GLEAMS) Available at: http://www.tifton.uga.edu/sewrl/Gleams/glm30pst.pdf Wilson RD, Geronimo J and Armbruster JA (1997) (non vide, referenced by NPTN) Available #2482314 File: 2006/1480/1

70

at : http://npic.orst.edu/factsheets/2_4-D.pdf Wolf M. (March 2001) European Commission, preview report for the active substance Amitrole directorate health & consumer protection. Available at: http://ec.europa.eu/food/plant/protection/evaluation/existactive/list1_amitrol_en.pdf Worthing C. R. (1983) the pesticide manual: A world compendium. Croydon, England: The British Crop Protection Council. (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryldicrotophos/dalapon-ext.html#20 WSSA (1989) Herbicide Handbook Committee (non vide, referenced by EXTOXNET (1996)) Available at: http://extoxnet.orst.edu/pips/amitrole.htm WSSA (1994) Weed Science Society of America Herbicide Handbook Seventh Edition Champaign, (non vide, referenced by EXTOXNET (1996)) Available at: http://extoxnet.orst.edu/pips/amitrole.htm WSSA Herbicide Handbook Committee (1989) Herbicide Handbook of the Weed Science Society of America, 6th Ed. Champaign IL (non vide, referenced by EXTOXNET (1993)) Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryldicrotophos/dalapon-ext.html#20 WSSA (2004) Imazapyr, Weed Control Methods Handbook, Weed Control Methods Handbook, the Nature Conservancy) http://tncweeds.ucdavis.edu/products/handbook/17.Imazapyr.pdf Xie L. Thrippleton K, Irwin MA, Siemering GS, Mekebri A, Crane D, Berry K and Schlenk D (2005) evaluation of estrogenic activities of aquatic herbicide and surfactants using a rainbow trout vitellegenin assay Toxicological Science Available at: http://toxsci.oxfordjournals.org/cgi/reprint/87/2/391.pdf

#2482314 File: 2006/1480/1

71

Appendices Appendix A: Herbicides usages during 2001/2002 and 2004/05 (after GMW) Herbicides

Control of weeds

Log Kow

Amount used (2001/02 figures)

Amount used (2004/05 figures)

Acrolein (a.i. 950g/kg)

Submersed weeds and algae (Ribbonweed, pondweeds, Elodea)

1.08

5445kg

794kg

2.58-2.83

10,100L

3284L

-0.97

28,666L

8618L