START: 01 JUL 2001 TERM: 30 JUN 2004 FY: 2004 GRANT YR: 2001
GRANT AMT: $127,711
INVESTIGATOR: Turechek, W. W.; Biggs, A. R.
PERFORMING INSTITUTION:
PLANT PATHOLOGY
N Y AGRICULTURAL EXPT STATION
GENEVA, NEW YORK 14456
REDUCED ANTIBIOTIC USE ON APPLES WITH REVISED MARYBLYT FORECASTING
NON-TECHNICAL SUMMARY: Fire blight is one of the most destructive and difficult-to-control diseases of apple. Over the last decade, major changes in horticultural practices have increased the chances for infection and level of damage likely to occur. A common approach to disease management in the Northeast is to time streptomycin antibiotic sprays during the blossoming period using the disease forecaster MARYBLYT. However, MARYBLYT does not account for varietal susceptibility, orchard age, or inoculum pressure (factors that may dramatically reduce (or increase) the risk of infection). As a result, disease management sometimes fails and, because outbreaks are so erratic, often results in unnecessary treatments when conditions do not support infection. While many growers are willing to forego the cost of possibly ineffective treatments as insurance against the potential losses of fire blight, unnecessary applications are costly and can lead to the appearance of resistant strains in the pathogen population. This project focuses on revising MARYBLYT, a computer forecaster for fire blight on apple and pear to account for varietal susceptibility, orchard age, or inoculum pressure.
OBJECTIVES: 1) Revise MARYBLYT to calculate a system of `risk points' as function of the MARYBLYT risk factors, then using the revisions, evaluate various management-action thresholds for fire blight disease management based on the accumulation of `risk points'; 2) Modify MARYBLYT to account for varietal susceptibility, orchard age, and inoculum potential; and 3) Develop a user-friendly, windows-based version of MARYBLYT.
APPROACH: MARYBLYT will be revised using a combination of historic and new data using a the statistical procedure 'receiver operating curve analysis'. The revised model will be field validated in unsprayed test blocks. The modifications will incorporate flexibility into MARYBLYT; allowing users to choose suitable action thresholds based on variety, orchard age, inoculum pressure, and their comfort for assuming risk. Ultimately, these changes will reduce the cost of managing fire blight by emphasizing efficient, and minimizing unnecessary, applications of antibiotics.
PROGRESS: 2001/07 TO 2004/06
A) During the course of the project, over 243 historical orchard data sets from
locations around North America and the UK were compiled and the predictive accuracy
of MARYBLYT (MB) was evaluated using receiver operator characteristic (ROC)
curve analysis. Historical data included climatic data during bloom, whether
or not disease occurred, and when symptoms appeared. MB uses 4 parameters to
predict blossom blight infection: 1) flowers open; 2) epiphytic inoculum potential
(EIP) exceeds 100; 3) rain or heavy dew within the last 24 hr; and 4) average
temperature greater than 15.6 C. Risk of infection is rated as low, moderate,
high, or infection depending upon whether one, two, three or all 4 of the thresholds
have been exceeded, respectively. Although infection is predicted only when
all 4 thresholds are exceeded, infections can occur in less favourable conditions,
i.e., when only 2 or 3 thresholds are exceeded. In practice, any combination
of those 4 thresholds could be used as a rule to trigger a management action.
ROC analysis permits us to evaluate each possible combination of thresholds
individually as a rule to trigger management action. For each data set, the
historical weather data was run through MB and, for each possible rule, it was
recorded whether infection was predicted or not, and the prediction was compared
to what actually occurred. For each rule, the true positive proportion (TPP),
defined as the number of true positives per number of data sets with disease,
and the false positive proportion (FPP), the number of false positives per number
of data sets without disease, were recorded. A perfect rule would have a ratio
between TPP and FPP of 1:0. The number of correct decisions when all the data
were combined was highest using the rule that predicts infection. The TPP:FPP
ratio was 0.73:0.41. The worst rule, 0.95:0.80, occurred when using temperature
and blossoming as the sole parameters for predicting fire blight. On the whole
the forecaster was predicting significantly better than chance (0.50) as is
shown by the 95 percent confidence interval (CI) of the area under the ROC curve.
The area was found to be between 0.61 and 0.73 and is an indication of the probability
of a correct decision. To investigate regional differences, the data were separated
into East Coast North America, West Coast North America and England. MB performed
best in England where the area under the ROC curve had a 95 percent CI of 0.55
to 0.88. Once again, the infection rule performed best with a TPP:FPP ratio
of 0.73:0.27. The worst performance was using the West Coast data where the
95 percent CI was 0.46 to 0.76 meaning that MB was not forecasting better than
chance. However, the TPP:FPP ratio for the West Coast was 0.65:0.32 when using
the infection rule. B) The software has been upgraded and converted into a Windows-based
format. Additions include: auto sensing for Fahrenheit or Celsius and a global
threshold option that will allow the operator to customize changes. Data is
now stored in an ASCII format and can be edited with any text editor software.
This eliminates the proprietary data storage structure of the old system.
IMPACT: 2001/07 TO 2004/06
A better understanding of the predictive abilities of both MB and CB has evolved
from this research. Scientists and consultants are learning the strengths and
weaknesses of both forecasters and can now use this information to better choose
forecaster and formulate management recommendations based on the forecaster
output. Moreover, the current results have directed our research focus to where
MB needs greatest improvement. The new windows version of MB is long anticipated
by the industry and should be well-received because of its ease in use as well
as its lower cost.
PUBLICATIONS: 2001/07 TO 2004/06
No publications reported this period
PROJECT CONTACT:
Name: NYG-Turechek, W. W.
Phone: 315-787-2474
Fax: 315-787-2389
Email: wwt3@cornell.edu