Environmental impact indices: DISCLAIMER: This presentation does not reflect the view of EFSA what do they reveal and not?

1. DISCLAIMER: This presentation does not reflect the view of EFSA
Environmental impact indices:
what do they reveal and not?
RR maize symposium: the European perspective
22-24 March 2010
22-
Yann Devos (PhD) – Junior Scientific Officer
GMO Unit – EFSA
Yann.Devos@efsa.europa.eu

2. 1. Introduction
Aim
– Assess and compare environmental impact of herbicide regimes
applied in genetically modified herbicide tolerant (GMHT) maize
with those used in its conventional counterpart
Residual
Residual + foliar
Residual Residual + foliar
GLY GLY
Residual GLY
Residual + GLY
2
sowing pre-emergence emergence 1 leaf stage; 2 leaf stage; 4 leaf stage; 5-6 leaf stage; 40 cm height 60 cm height
3-4 cm height 3-4 cm height 4-10 cm height 10-15 cm height

3. 1. Introduction
Environmental impact indices
– Pesticide Occupational and Environmental Risk Indicator
(POCER) → Vercruysse & Steurbaut (2002)
• Maize: Devos et al (2008)
– Environmental Impact Quotient (EIQ) → Kovach et al (1992)
• Maize: Leroux et al (2006); Kleter et al (2007); Brookes & Barfoot
(2008)
• Soybean: Kleter et al (2007); Bonny (2008); Brookes & Barfoot
(2008)
• Cotton: Brookes & Barfoot (2008)
• Oilseed rape: Brimner et al (2005); Kleter et al (2007); Brookes &
Barfoot (2008)
3

4. 2. POCER → Vercruysse & Steurbaut (2002)
Pesticide Occupational and Environmental Risk Indicator
(POCER) – modules
– Annex VI of Directive 91/414/EEC
• 3 modules for human health (non-dietary exposure)
– Risk to pesticide operator
– Risk to worker
– Risk to bystander
• 7 modules for the environment
– Persistence in soil
– Risk of ground water contamination
– Acute risk to aquatic organisms
– Acute risk to birds
– Acute risk to bees
– Acute risk to earthworms
– Risk to beneficial arthropods 4

5. 2. POCER → Vercruysse & Steurbaut (2002)
Pesticide Occupational and Environmental Risk Indicator
(POCER) – formula
– For each module → risk is estimated via risk indices (RI)
Risk index Estimated exposure / toxicity ratio
Pesticide operator IE / AOEL [IE=internal exposure; AOEL=acceptable operator exposure level]
Worker (DE x AbDE) / (AOEL x BW) [DE=dermal exposure; AbDE=dermal absorption
factor; BW=body weight]
Bystander (DE x AbDE + I x AbI) / (BW x AOEL) [I=inhalation exposure]
Persistence 10[((DT50/90)-1) x 2] ) [DT50=half-life]
Groundwater PEC / 0.1 [PEC=predicted environmental concentration in groundwater]
Aquatic organisms PEC / MTC [MTC=maximum tolerable concentration]
Birds (10 x PEC) / (LC50 x BW)
Earthworms (10 x PEC) / LC50
Bees AR / (50 x LD50) [AR=application rate]
5
Beneficial arthropods (RC – 25) / (100 – 25) [RC=reduction of control capacity]

6. 2. POCER → Vercruysse & Steurbaut (2002)
Pesticide Occupational and Environmental Risk Indicator
(POCER) – calculations
– Integration of RI into total risk indicator
• Describe extent to which a chosen trigger is exceeded as a
numerical dimensionless value
– Step 1 – define lower (LL) and upper limit (UL) for each RI
– Step 2 – calculate relative RI, LL and UL & log-transform
– Step 3 – determine exceedence factors (EF)
» EF values ≤ 0 are scored as 0 → low risk
» EF values ≥ 1 are scored as 1 → high risk
» EF values between 0 and 1 → intermediate risk
– Step 4 – calculate total risk = ∑ EF values ranging between 0 and 10
» Assumption: all components are equally important
6

7. 2. POCER → Devos et al (2008)
Herbicide regimes in conventional maize
– 3 different strategies to control annual/perennial grasses and
broadleaf weeds (abbreviated as CON)
• Pre-emergence of crop
• Early post-emergence, ideally in 2-4 leaf stage of maize
• Sequentially,
– where a combination of herbicides with soil (residual) activity is
applied pre-emergence
– followed by a mixture of post-emergence herbicides with foliar
activity
– Farmers use a combination of <3-4> active substances
– 13 typical herbicide regimes (Flanders; Belgium)
• Time of application; dose; activity; weed spectrum
7

8. 2. POCER → Devos et al (2008)
Herbicide regimes in RR maize
– Different strategies to control annual/perennial grasses and
broadleaf weeds (e.g., Dewar, 2009)
• Single or sequential application of GLY only, without relying on
pre-emergence herbicides
• Use of GLY in combination with other herbicides, especially
residual herbicides applied pre-emergence
• Use of GLY in a single application in combination with other
post-emergence herbicides with residual activity
– 10 GLY-based herbicide regimes
• Single vs. sequential application; dose; application timing;
presence/absence of residual herbicide
• RR composition = 360 g/l
8

9. 2. POCER → Devos et al (2008)
Herbicide regimes in RR maize
– 3 regimes:
• Single application of GLY only (abbreviated as GLY)
• Application dose rates (g/ha):
– 720 → medium efficacy (Soukup et al, 2008)
– 900 → medium efficacy (Leroux et al, 2006)
– 1080 → medium efficacy (Phipps and Park, 2002)
– 4 regimes:
• Sequential application of GLY only (abbreviated as GLY)
• Application dose rates (g/ha)
– 900 + 450 = 1350 → high efficacy (Leroux et al, 2006)
– 720 + 720 = 1440 → high efficacy (Monsanto)
– 900 + 900 = 1800 → high efficacy (Leroux et al, 2006; Monsanto)
– 1080 + 1080 = 2160 → high efficacy (Soukup et al, 2008;
Monsanto) 9

10. 2. POCER → Devos et al (2008)
Herbicide regimes in RR maize
– 3 regimes:
• Single application of GLY in combination with herbicides with
residual activity (abbreviated as GLY+)
• Application dose rate (g/ha)
– GLY (1080) + acetochlor (2100) → high efficacy (Soukup et al, 2008;
Monsanto)
– GLY (1080) + herbicide with residual activity (full dose rate)
» S-metolachlor
» Terbuthylazin
» Dimethenamid-P
10

11. 2. POCER → Devos et al (2008)
Results – human health
– 3 modules
• Risk to pesticide operator
– EF CON = [1.00]
– EF GLY = [0.54-0.78]
– EF GLY+ = [1.00]
• Risk to worker
– EF CON/GLY/GLY+ = [0.00-0.37]
• Risk to bystander
– EF CON/GLY/GLY+ = [0.00]
– If used alone, GLY has lower impact on pesticide operator than
other herbicide regimes tested
– Risk to worker and bystander is low and transient
11

12. 2. POCER → Devos et al (2008)
Results – environment
– 7 modules
• Persistence in soil
– EF CON/GLY/GLY+ = [0.00-0.03]
– Half lives ≤ 90 days considered low
• Risk of ground water contamination
– EF CON/GLY/GLY+ = [0.18-0.33]
– Risk of ground water contamination low due to rapid adsorption in
soil of GLY
• Acute risk to aquatic organisms
– EF CON = [0.47-1.00]
– EF GLY = [0.00]
– EF GLY+ = [0.38-1.00]
– GLY has low acute toxicity to fish, Daphnia and algae
12

13. 2. POCER → Devos et al (2008)
Results – environment
– Acute risk to birds / bees / earthworms / beneficial arthropods
• EF CON/GLY/GLY+ = [0.00]
• Low acute toxicity to birds, bees, earthworms and beneficial
arthropods
Overall conclusion
3,0
Exceedence factor (EF)
POCER modules
2,5
2,0 CON
values
GLY
1,5
GLY+
1,0
0,5
0,0
13
Human health Environment Total

14. 3. EIQ → Kovach et al (1992)
Environmental Impact Quotient (EIQ) – components and
calculations
14

15. 3. EIQ → Leroux et al (2006)
EIQ-methodology applied to RR maize in Canada (Québec)
15

16. 3. EIQ → Kleter et al (2007)
EIQ-methodology applied to GMHT crops in US
– 2004; pesticide use survey data of National Center for Food and
Agricultural Policy (NCFAP); percent change
– Proportional EIQ/ha reduction of 39% in maize
Soybean Ecology impact, EI/A
Maize Consumer impact, EI/A
Cotton Farmworker impact, EI/A
Total impact, EI/A
Canola
Pesticide use, lbs ai/A
0 20 40 60 80
% decrease GM vs. conventional
16

17. 3. EIQ → Brookes & Barfoot (2008)
EIQ-methodology applied to GMHT maize globally
– 1997-2006; pesticide use survey data from US, Canada, South
Africa & Argentina
17

18. 3. EIQ → Bonny (2008)
EIQ-methodology applied to GMHT soybean in US
– 1990-2006; pesticide use survey data of US Department of
Agriculture (USDA)
Period Field EIQ
1994-1996 29.2
2001 20.4
2002 23.8
2006 25.7
18

19. 3. EIQ → Brookes & Barfoot (2008)
EIQ-methodology applied to GMHT soybean in Romania
– 2000-2003; data from Brookes (2005)
19

20. 4. What environmental impact indices do not
reveal?
– (see e.g., Cerdeira & Duke, 2006, 2010; Dewar, 2009, for
comprehensive reviews)
• Weed control efficacy & weed management flexibility
• Impact due to the adoption of conservation tillage practices
• Impact on human health due to pesticide residues
• Impact of GLY metabolites (e.g., AMPA)
• Risk to mammals
• Weed resistance evolution to GLY
• Weed spectrum shifts
• Impact on farmland biodiversity
• Impact on microorganisms and soil functions
• …
20

21. 5. What environmental impact indices do
reveal?
Useful tools
– as indicators of environmental impact of pesticides
– to compare/rank pesticides based on environmental impact
Herbicide regimes in maize cropping systems
– GLY-based herbicides have a better environmental profile
compared to those applied in conventional maize
– Addition of herbicides other than GLY in RR maize
reduces/cancels beneficial effect, depending on application
dose rate of additional herbicide
21

22. 6. Thank YOU for your attention!
Acknowledgments
– Dirk Reheul & Mathias Cougnon & Robert Bulcke
• University of Ghent; Department of Plant Production
– Sofie Vergucht & Walter Steurbaut
• University of Ghent; Department of Crop Protection
– Geert Haesaert
• University College of Ghent; Department of Plant Production
– Gijs Kleter
• RIKILT; Institute of Food Safety; Wageningen University and
Research Center
22