A decision Support System to adjust herbicide use to field conditions Research Centre Bygholm 14 Nov

1
A decision Support System to adjust herbicide use
to field conditionsResearch Centre Bygholm 14
November 2005

• Per Rydahl
• Danish Institute of Agricultural Sciences (DIAS)
• Dept. of Integrated Pest management
• Research Centre Flakkebjerg, Denmark

2
Disposition

• Models and algorithms
• Validation and implementation
• Demonstration
• Field- and sitespecific use

3
Models and algorithms
4
Focus

• Low input of herbicides
• reduce environmental loadIn Denmark political
  action plans for reductions on the Treatment
  Frequency Index (TFI)
• reduce costs

5
A 3-step model function

• assesses the level of weed control needed
• selects single herbicides and calculates doses
  needed
• calculates tank-mixtures, optimised for cost or
  TFI

6
Model step 1The level of control needed

• Includes aspects on yield, yield quality and
  long term weed control
• Based on expert knowledge
• Output the level of control needed on weed
  biomass, 4-6 weeks after a herbicide application

7
Model step 2 Dose-response function, 3
herbicides
8
Step 2 Dose-response function 1 herbicide, 1
weed species (low susceptibility)
1/4 N 1/2 N 1/1 N 2/1 N
9
Model step 2 1 herbicide, 1 weed species (high
susceptibility)
1/4 N 1/2 N 1/1 N 2/1 N
10
Model step 2 Fitting data to model
11
Model step 2 1 herbicide, 3 weeds
12
Step 2 Dose-response function 1 herbicide, 1
weed, 4 growth stages
13
Model step 2 Dose-response function, general
attributes
14
Model step 2 Dynamics of dose-response function
15
Model step 3Tankmixtures

• Motives
• Reflect practical experiences
• Extend weed spectrum
• Save costs or TFI
• Method
• Additive Dose Model (ADM)
• linear optimisation of 2-4 component mixtures

16
Validation and implementation
17
Model validationSpring cereals
Prototypes 702 tests, 1987-98 Present DSS
12 tests, 1996-97, 263 weeds per m2
18
Model validationSugar beet
Prototypes 42 tests, 1995-2000 Present CPO
8 tests, 1999-2000, 80 weeds per m2
19
Status on implementation

• Denmark
• 12 major crops
• All registered and marketed herbicides
• 75 weed species
• subscribers
• 1000 farmers, 300 consultants
• all relevant schools, companies etc.
• System export
• Estonia, Latvia, Lithuania, Poland and Norway

20
Major obstacles

• DSS development
• availability of on the biological activity of
  herbicides in different dose rates for different
  conditions
• DSS implementation
• field registrations before spraying

21
Conclusions (1)

• gt2000 Danish field tests
• no adverse effects on yield
• no adverse effects on residual weeds
• Herbicide input in single fields TFI 0,05 -
  gt2,00
• Compared to sales statistics, average cost/TFI
  of herbicides was reduced
• Cereals gt50
• Sugar beet about 20
• Peas and oilseed rape insignificantly
• Maize (tests not yet completed)

22
Conclusions (2)

• In the context of CPO, a ranking of factors
  affecting the potential for reducing input of
  herbicides can be made
• the level of weed control needed
• weed species
• weed growth stages
• herbicide mixing
• temperature and Rh
• water stress (rare in Denmark)

23
Demonstration
24
Solve weed problem - input
www.planteinfo.dk
25
Solve weed problem - output
26
Field- and site-specific use
27
Field-specific and regional use of CPO

• Case study
• detailed weed registrations in 26 fields of
  winter wheat (gt200 weeds/m2 on average)
• single field use of CPO
• 5 weed species per field in 26 fields
• TFI 0,25
• 2-5 fields in 6 regions
• 11 weed species per region
• TFI 0,50
• 26 fields in 1 region
• 25 weed species
• TFI 1,02

28
Ideas for site-specific use of CPO

• Weed mapping
• step 1 register weeds
• step 2 produce weed map and application map
• step 3 spray according to application-map
• Real-time
• registration of dominant species, densities and
  growth stages
• calculation of treatment options with CPO
• spraying

29
Developments needed forsite-specific use of CPO

• automatic registration of weed species, weed
  densities and weed growth stages
• analyses on the accuracy required in weed
  registrations to match sensitivities in CPO
• integration of
• automatic registration
• CPO-calculations
• spraying systems

30
Thank you for your attention!