Disease Forecasting: TOMCAST as a case study PP 603

1
Disease ForecastingTOMCAST as a case
study PP 603

• Jim Jasinski
• OSU Extension
• Integrated Pest Management Program

2
Outline of Topics

• Industry Overview
• Disease model basics
• TOMCAST
• History
• Research
• Equipment
• Exercise
• Weather / disease forecasting networks
• OWN / WIN, NA Disease Forecast Center, MAWN
• SkyBit website and resources

3
OH Tomato Industry Overview

• Whole pack, paste, fresh market
• Ohio Ag statistics proc. tomato
• From 1999 - 2008, 13 M - 18 M
• From 1999 2008, 17,700 - 5,900 A
• Ohio Ag Statistics fresh market
• From 1999 2008, 28 M - 61 M
• From 1999 2008, 4,900-6,000 A
• 2007 OMAFRA
• 18,000 A of processing tomatoes in Ontario
• ca. 74 M

4
OH Processing Tomato Industry
5
Classic Disease Triangle
Disease
6
Classic Disease Triangle Time
Environment
Pathogen
Host
Time
7
What is a Disease Forecasting Model?

• A set of formulas, rules, tables, or algorithms
  patterned after the biology of a specific
  pathogen
• Models are driven by observed or forecasted
  weather conditions for each location
• air temperature
• relative humidity
• leaf wetness (dew)
• precipitation
• others

8
Why Use Forecasting Models?

• Alternative to calendar spray programs
• Enhance timing of fungicide sprays based on
  disease development
• Spray reduction may be possible
• Economic benefits
• Environmental benefits

9
3 Stages of a Model

• Development
• Assumptions monitoring variables
• Validation
• Testing the assumptions
• Implementation
• Grower trials, release to public

10
1. Model Development

• Models typically are developed from a
  combination of laboratory and field studies
• The goal is to predict the risk of disease and/or
  development of inoculum
• Need to identify key environmental and host
  variables
• Air temperature, relative humidity, hours of free
  moisture (dew), precipitation, host growth
  stages, etc.
• Based on management options and goals, action
  thresholds can be incorporated into the model to
  provide advice on fungicide applications

11
2. Model Validation

• Descriptive models must be validated across a
  variety of microclimates over a number of years
• Pilot studies (multiple locations)
• Multiple seasons (hot, cold, wet, dry weather)
• Sensor placement (canopy v. field edge?)
• Revise and refine based on results
• Models developed in one area are frequently
  validated by researchers in other areas may need
  region-specific modifications
• Plants treated according to the model are
  compared to disease levels managed by traditional
  spray schedules as well as unsprayed plots

12
3. Model Implementation

• Predictive models require local weather inputs
• On-site data logger or access to forecasted data
• air temperature, RH, precipitation, etc.
• Quality data very important (GIGO)
• Initial implementation efforts are supported by
• Industry, University researchers, Extension
  agents
• Through field days, demonstrations, and on-farm
  trials
• Programs can be transferred from public
  (University) to the Private domain (consultants /
  industry / subscription service) for maintenance
• Growers and pest control advisors can use disease
  risk indices (DSV) for enhanced crop management

13
More Models - UC Davis IPM
http//www.ipm.ucdavis.edu/GENERAL/tools.html
14
Online disease forecasting models
http//www.ipm.ucdavis.edu/DISEASE/DATABASE/diseas
emodeldatabase.htmlDISEASEMODEL
15
Late blight forecasting models
16
Late blight forecasting models
17
Late blight forecasting models
18
Disease Forecasting Models/Networks

• EPIDEM- Alternaria solani on tomatoes potatoes
• FAST - Forecasting Alternaria solani on tomatoes
• TOMCAST-Alternaria, (Septoria, anthracnose)
• WISDOM (BLITECAST)- Late blight on tomatoes
  potatoes
• MELCAST- Anthracnose, gummy stem blight
  (Watermelons), Alternaria (Muskmelons )
• Maryblight- Fireblight on apples
• EPIVEN- Apple scab on apples
• EPICORN- Southern corn leaf blight
• North American Blue Mold warning system- Tobacco
• North American Plant Disease Forecast
  Center-Cucurbit Downy Mildew, Soybean Rust,
  Tobacco blue mold
• Weather Innovations Incorporated DONcast,
  SPUDcast, TOMcast, BEETcast
• Michigan Automated Weather Network-TOMCAST,
  Potato late blight, others

19
TOMCAST TOMato disease foreCASTing

• Purpose Assist processing tomato growers with
  fungicide application timing based on early
  blight development, using a protectant
  fungicide program
• Use local weather to guide fungicide schedule
• Alternative to 7-14 day calendar spray programs
• Only a PART of the Disease management component
  of an overall IPM Program for tomatoes

20
TOMCAST Disease Spectrum

• PREDICTED

Early Blight
Clemson University

• NOT PREDICTED
• -Viral (TSWV), Bacterial (Canker, Speck, Spot),
  Fungal (Late Blight??)

21
Factors Affecting Early blight

• Early blight spores need free moisture to
  germinate and infect tomato plant tissue. If the
  leaf surface is dry, the fungal spore just sits
  there...and waits.
• The rate or speed of infection is determined by
  temperature. The cooler the temperature the
  slower the process whereas the warmer the
  temperature the faster infection takes place,
  disease occurs.
• An environmental factor (moisture) starts the
  disease and another environmental factor
  (temperature) speeds up or slows down the
  progress of the disease.

22
Alternaria biologyEnvironmental Conditions for
Germination Sporulation

• Alternaria spores
• Germinate within 2 hours over a wide range of
  temperatures but at 80 to 85oF may only take 1/2
  hour.
• Another 3 to 12 hours are required for the fungus
  to penetrate the plant depending on temperature.
  
• After penetration, lesions may form within 2-3
  days or the infection can remain dormant awaiting
  proper conditions (60oF and extended periods of
  wetness).
• Alternaria sporulates best at about 80oF when
  abundant moisture is present (rain, mist, fog,
  dew, irrigation). Infections are most prevalent
  on poorly nourished or otherwise stressed plants.

23
Protectant Fungicide Activity
Protectants
Germination Penetration Infection
Sporulation
Chlorothalonil EBDCs
Strobilurins
NA for toms
24
TOMCASTDisease Severity Value (DSV) Chart

• TOMCAST DSV generated on 24 hour intervals
• Calculation requires leaf wetness and air
  temperature inputs
• Count leaf wet hours, average temperature
  during those leaf wet hours, determine the daily
  DSV

25
TOMCASTDisease Severity Value Chart

• 0-Environ. conditions unfavorable for disease
  development (g,p,i,s)
• 4-Environ. conditions highly favorable for
  disease development

26
TOMCAST Math

• Add daily DSV until spray threshold is reached or
  exceeded
• 15 20 DSV is the usual spray threshold or spray
  interval
• Example
• Monday 10 lw hrs 65 F 2 DSV
• Tuesday 7 lw hrs 75 F 2 DSV
• Wednesday 13 lw hrs 72 F 3 DSV
• Thursday 9 lw hrs 69 F 2 DSV
• Friday 16 lw hrs 78 F 3 DSV
• 5 days 12 DSV

27
TOMCAST History

• Original Model- FAST
• Forecasting Alterneria solani on Tomatoes
• Pennsylvania State College, Late 70s
• Air Temp Wetness variables
• Rain and Dew models
• DSVs averaged 5 or 7 days
• BLITECAST is born
• NY CU-FAST
• Cornell Univ. (1989)
• Sold CU-FAST
• Privatized (1992)
• charge fee for use

PENNSYLVANIA NEW YORK
28
TOMCAST History

• 1979 Ontario, Canada
• FAST system unacceptable (complex, hardware)
• 1985
• TOMCAST is born (R. Pitblado)
• Used datalogger w/ new dew sensor
• Daily DSV calculations
• 1987
• Industry University Support
• Grower trials begin
• 2000-present Ontario Weather Network became
  Weather Innovations Inc.,Tomcast, Beetcast, etc.

CANADA
29
TOMCAST History

• 1988 TOMCAST arrives in Midwest
• University Industry supported research
• Heinz, Campbells, Hirzel
• 1992 Grower Trials
• Datapod Recorders
• 1993 Upgraded to CR10 Equipment
• 1995 Network expands
• BLITECAST, Fresh Market, Whole pack, Skybit
• 1999 14 CR10 stations, mature research service
  program
• 2000 Program review - proc. tomatoes decline (5K
  A), TOMCAST discontinued in Midwest

30
TOMCAST in Midwest

• Research (Development Validation)
• 1988-1991
• Split plot research on commercial farms
• Disease, insect, yield data collected
• Spray program of TOMCAST v. Calendar (UTC?)
• Can reduce chemical inputs (25-40)
• Produces quality or gt calendar program
• Extension Service (Implementation)
• 1992-1999 Recruit growers expand network (14
  sites)
• Educational focus, how to use, support, data
  dissemination
• Continue research aspect, Late blight, SkyBit,
  Whole pack

31
TOMCAST Research Design
TOMCAST
or
Conventional

• Scout fields weekly
• Harvest plots

32
TOMCAST EquipmentEarly Years...

• Omnidata Datapods
• Simple dataloggers
• Manually read (15-45 min. daily)
• EPROM chips
• Data manually recorded calculated
• Reliability issues
• Labor intensive
• Relatively Cheap

33
TOMCAST EquipmentLater years...

• Campbell Scientific CR10
• Programmable data loggers
• Very Reliable
• Very low maintenance
• Collection via phone lines and modem (now
  wireless possibilities)
• One person can run the entire data collection
  network
• Automatic logging / interrogation of units via
  telecommunications software
• Expensive (2500 / unit)
• Future is web based collection dissemination

34
CR10 in the field
35
How the leaf wetness sensor functions
Electrified Copper Grid

• Latex paint coated sensor
• Dry, max resistance, no conductivity
• Increasing moisture, lt resistance,
• gt conductivity

36
Leaf wetness sensor

• Latex paint coated sensor

37
CR10 Guts
1.2 Kb modem
38
TOMCAST Network Data Retrieval Loop
modem
modem
CR10
central office
field
39
Web Access for Growershttp//vegnet.osu.edu/tomca
ts/tomfrm.htm
40
TOMCAST DSV Accumulation Midwest Use Guidelines

• Initial Spray based on transplanting date
• Before May 20th - 25 DSV or fail safe of June 15
• After May 20th - Use Spray threshold (15-20 DSV)
• Successive sprays based multiples of DSV spray
  threshold.
• 15 DSV, more conservative (gt sprays)
• 20 DSV, less conservative (lt sprays)
• Accumulate Daily DSV until threshold reached
• 14 day failsafe spray interval

41
DSV Accumulation Ex. - Slow
July
14 day spray needed fungicide degradation and
new foliage
spray
42
DSV Accumulation Ex. Rapid
Rapid foliage fruit growth require increased
sprays for coverage
Spray
43
TOMCAST DSV AccumulationWork Sheet Guidelines

• Guidelines
• First Spray ONLY!
• Transplant before May 20th - 25 DSV or fail safe
  of June 15
• Transplant after May 20th - Use Spray threshold
• Successive Sprays
• Calculate sprays at 7 and 10 calendar days
• Calculate sprays at 15 18 DSV
• Spray at 14 Days if threshold not reached (DNA
  1st spray)

44
Forces Driving TOMCAST

• Limited Fungicide Selection
• Prior to 1996 used Delaney clause standard
• No (zero) pesticide residues of carcinogenic
  compounds can be found in processed foods
• Eliminated use of EBDCs (Ethylene
  bisdithiocarbamate) because of ETU (Ethylene
  thiourea) metabolite
• Thiram, Mancozeb, Ferbam, Ziram, and Zineb
• Workhorse is Chlorothalonil
• After 1996 Delaney revoked, used FQPA
• Standard reasonable certainty of no harm
• EBDCs can be used (see above)
• New chemistries Strobulurins (Quadris, Flint)
• Effect on DSV action thresholds?

45
Network Geography Topography

• Zones of DSV prediction?
• Flat land
• ca. 15 miles
• Lake effect
• ca. 5-10 miles?
• Hilly
• ca. 5 miles?

Weather Station or grower

46
Where is Ontario, eh?
47
OWN Home Page
48
OWN Becomes WIN
49
Weather Innovations Inc.
50
WIN TOMCAST Maps
51
WIN TOMCAST Maps
52
WIN TOMCAST Maps
53
WIN TOMCAST Maps
54
MI Automated Weather Network
55

56
MI Automated Weather Network
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
MI Automated Weather Network
61
MI Automated Weather Network
62
MI Automated Weather Network
63
MI Automated Weather Network
64
MI Automated Weather Network
65
(No Transcript)
66
(No Transcript)
67
(No Transcript)
68
Project spore path of non-overwintering diseases
(in the North), such as downy mildew from NC
through the east coast in 54 hours based on
weather transport models.
69
SkyBit Inc.Subscription Forecasted Weather

• Alternative to data loggers?
• No Equipment Maintenance
• Hardwareless
• A commercial company
• Product is localized forecasted
• agricultural weather
• Target IPM Programs / Models
• Uses satellite, NWS, radar and remote sensing
  data
• Resolution 1 sq. Km
• Available by subscription (ca. 50 / product /
  field)

70
SkyBit Home Page
71
SkyBit Product Examples
72
(No Transcript)
73
(No Transcript)
74
(No Transcript)
75
(No Transcript)
76
(No Transcript)
77
SkyBit CR10 Research-Research Sites
1995-97-Compared temperature, leaf wetness,
rate of DSV accumulation-5 locations
78
DSV Accumulation GraphCR10 vs. SkyBit
TOMCAST
SkyBit
Consequences of over or under DSV estimates?
79
SkyBit v. CR10 Research Highlights

• Leaf wet hours
• SkyBit forecasted higher temps than CR10
• Result- increased SkyBit DSV
• Non-leaf wet hours
• SkyBit forecasted lower temps than CR10
• Result- no significant effect
• SkyBit forecasted fewer leaf wet hours per day
  compared to CR10 observations
• Result- lower SkyBit DSV accumulation

80
In Conclusion

• Forecasted weather products and area wide weather
  networks are becoming prevalent
• Disease models exist for fruits, vegetables,
  field crops, and turf
• The use of predictive models can help growers
  better manage disease in their crops
• The use of predictive models will increasingly be
  a part of an overall IPM program