PestCast banner
Disease Model Database

TomatoPM: A Spray Forecast Model of Tomato Powdery Mildew
by R. A. Guzman-Plazola, R. M. Davis, and J. J. Marois, University of California, Davis, in conjunction with the UC Statewide IPM Program

What is TomatoPM?
Typical use of TomatoPM
Data collection for input into the model

Data file preparation
Run model and update spray forecast
Bugs in this disease program

Before using TomatoPM: This model is based on research sponsored by the California Tomato Research Institute and California Tomato Board and conducted by R. M. Davis, R. A. Guzman-Plazola, and J .J. Marois from 1993 through 1996. The model is based on results from experiments under controlled conditions (growth chamber) and field monitoring of disease and weather data. A linear discriminant function was generated to classify daily environmental conditions of tomato canopies according to their conduciveness for disease. Cross-validation of the discriminant function and tests with independent data sets yielded percentages of correct classification of 74-96% and 87%, respectively.

The model was tested during two consecutive years (1995 and 1996) in ten tomato fields (fresh market and processing) in the northern San Joaquin and southern Sacramento valleys of California. After following the recommendations of the model, it was possible to save 2 to 5 fungicide sprays with no significant impact on fruit yield and quality. Disease levels at harvest time in plots managed according to the model have been equal to or only slightly higher than that of plots where fungicide applications were made every 14 days.

While these results are very promising, remember that

  • Sensors must be calibrated to give accurate results.
  • The model uses microclimate data from tomato canopies, so proper placement of the sensors is very important for correct classification of daily conditions.
  • If using the model in new areas, it should be tested for one or more seasons under local conditions to verify that it will work in your location. Monitor fields closely for disease while testing the model.

You can download TomatoPM, a working version of this model prepared for use with Microsoft Excel 5 and 97/98 for Windows or for Macintosh. (Add Excel as a helper application in your browser for files with extension ".xls" before downloading.) Below is information to help you collect the required data and use it to run the working version of the model.

Hardware and software requirements. This working version requires Microsoft Excel 5 or 97/98, running under Windows 3.1 or 95 or on a Macintosh computer. As this initial version is improved, new versions will be posted here. Send e-mail to webmaster to get on a mailing list for future updates.

What is TomatoPM?

TomatoPM is a computer representation of the spray forecast model for tomato powdery mildew developed by R. Guzman-Plazola, R. M. Davis, and J. .J. Marois at University of California, Davis. This computer version was developed by Joyce Fox Strand, Statewide IPM Program, UC Davis.

TomatoPM is an Excel 5 workbook (file) which includes a prepared worksheet and macros that

  • Process the hourly data you have collected.
  • Classify the conditions for each day in terms of conduciveness for disease.
  • Evaluate the daily conditions over the most recent period (usually 6 days) to assess the risk of low, moderate, or severe disease.
  • Recommend management action based on the conditions over the period.

The TomatoPM spray forecasts and the model calculations for each day can be printed, and the daily calculations can be used to create charts or for further analysis.

Typical use of TomatoPM

  1. Collect the weather data from the tomato field.
  2. Insert the data onto a worksheet in an Excel 5 workbook (Excel 5 name for its file of spreadsheets).
  3. Run the Excel 5 model to update the forecast using the most recent weather data.

Data collection for input into the model

Required sensors.
One temperature (T), one relative humidity (RH), and one leaf wetness (LW) sensor are required. If two sets of sensors are used, the data for each set will be averaged. If more than two sets are used, hourly values for all of them should be averaged before using TomatoPM. The averages should then be placed in the data columns for Sensor 1, with Sensor 2 left blank.

Placement of the sensors. Proper placement and calibration are critical for reproducing the research results.

T and RH. Each set of sensors for T and RH should be kept together (they are usually parts of a single probe) and protected by a shield to avoid direct exposure to the sun and water, but the shield must allow good air flow around the sensor.

Since the forecast model uses microclimatic data at the canopy level, T and RH sensors must be placed inside the foliage and toward the top of the plants. Location of the sensors must be checked regularly, at least once a week, particularly during the fast growth stages of the tomato crop and after periods of high speed winds.

LW. Leaf wetness sensors must be placed on the top of the canopy and relocated periodically according to the dynamics of plant growth.

Output variables. The datalogger should make several measurements from each sensor during an hour, to be averaged over the hour. Measurements reported more often than each hour must be summarized into hourly values for input into this working version of the model.

The data values required by the model, and their units, are listed in the following table.


 Units/Expected Values

 Data Spreadsheet

 Julian day

 Day of year from 1 to 365
(366 if leap year)


 Hour of the day

 0-2300, or 100-2400


 Average of T for each hour, Sensor 1



 Average of T for each hour, Sensor 2



 Average of RH for each hour, Sensor 1



 Average of RH for each hour, Sensor 2



 Leaf wetness, Sensor 1

 Scale 1-10


 Leaf wetness, Sensor 2

 Scale 1-10


 Maximum T for each hour, Sensor 1



 Maximum T for each hour ,Sensor 2



 Minimum T for each hour, Sensor 1



 Minimum T for each hour, Sensor 2



 Maximum RH for each hour, Sensor 1



 Maximum RH for each hour, Sensor 2



 Minimum RH for each hour, Sensor 1



 Minimum RH for each hour, Sensor 2



Data file preparation

TomatoPM expects to find the data in a worksheet in an Excel 5/97-98 workbook. You may download a sample data file "stockton.xls" (for Windows or Mac) to use to try out TomatoPM. Data in this file start at Julian day 225 and end on day 267.

The worksheet should be in the following format:

Cell B1: Name for the location of the weather station
Starting in Row 3, columns A through P should include an hourly value of the variables as listed in the above table, for at least Sensor 1 of each pair.

If you have only one sensor for T, RH, or LW, TomatoPM expects to find the data in the column labeled for sensor 1 and the sensor 2 column left BLANK.

All hours must have a value. If you have missing data, fill in from hours on a similar day, use data from a similar station, or in some other way fill in the gaps. If you cannot fill the gaps, you will need to start a new sheet and enter a starting date that falls after the gap as the beginning date.

Run the model and update the spray forecast

TomatoPM prepares one worksheet for each field/weather station location you plan to enter. The program then takes the data you have prepared in the data workbook and uses it to generate a forecast on the highlighted area of the TomatoPM worksheet.

The buttons to the left of the highlighted area of the TomatoPM worksheet allow you to:



Add New Sheet

add new locations (each is a worksheet). Note that the file will become large with lots of locations, and processing may be slow.

Delete This Sheet

delete unwanted locations (sheets)

Add Data & Update Forecast

have TomatoPM search the data worksheet for new data, then update the forecast using the new weather data

Enter Spray & Update Forecast ONLY

enter a spray, even without new weather data, and update the forecast using the spray information

Review Sprays

see which dates have been designated as days on which sprays were made

View Data for Analysis

View the daily forecast information for the season. Also, copy from here to put into another spreadsheet for further analysis or graphing.

Print Forecast

print the forecast display

Print Data

print the daily information in spreadsheet form

Getting started: set up a location sheet

Open the TomatoPM workbook in Excel. Note that one of the sheet tabs on the bottom of the screen says Set_up. This worksheet is the main template for the model. To generate forecasts, you must ADD a new sheet for each location for which you are generating forecasts.

  1. Open TomatoPM.
  2. Press the Add Sheet button on the Set_up sheet.
  3. When asked, enter a name for the field/weather station location (Davis W1, for example).
  4. If you have weather stations in multiple locations and you will be generating forecasts for each, add a new sheet for each location.

A new worksheet will be created and labeled with the location name you supplied. This is the worksheet on which the forecast will be generated for data from that location.

Updating a new forecast from new data

  1. Select the location sheet you wish to update.
  2. Press Add Data & Update Forecast.
  3. When asked, indicate
    • the Julian day of the first day of data to add (first time only)
    • the Julian day of the anticipate harvest date for the field (first time only)
    • the number of days to add
    • the name and location of the data workbook (first time only)
    • the name of the data sheet in the data workbook (first time only)
    • the Julian date of any fungicide sprays you have applied

TomatoPM will search the data workbook/worksheet for values for the requested days, copy the data to the TomatoPM location sheet, calculate the daily values and values for the evaluation period, then display the forecast, recommendations, and a data summary in the highlighted area on your screen.

Interpreting the forecast

The forecast display includes the following information:

Last day of evaluation period

"Day" is the last day of data used in the evaluation. Both Month/Day and Julian day are presented, and the estimated harvest date is shown.

Recommended action

From the disease risk evaluated for the most recent period, an action (spray, don't spray, reevaluate) is recommended.

Disease risk

No or low disease, moderate disease, or severe disease is expected, based on conditions over the most recent period.

Summary for most recent 10 days

Date (and Julian day) for each day in the most recent period (usually 6 days), and the evaluation of the weather conditions for each day: Conducive, Moderate, Nonconducive. Spray recommendations are shown in red type, sprays made are in bold.

Data entered from (day) to (day)

Dates indicating the total period for which data have been added to this location sheet.


The name of the location, from the sheet tab.

Today is (date)

Current date shows when the forecast was generated.

Press the Print Forecast button to print a copy of the forecast display. Press the Print Data button to print a copy of the detailed daily and period evaluations.

Note: The spray forecast and recommended action provide you with additional guidance to use in making a decision about when (and if) to spray a fungicide to control powdery mildew in your field. The applicability of the forecast to fields where no weather data are being collected is unknown; reliability will depend on the similarity of micrometeorological conditions between the field where the weather is monitored and other fields to which you are attempting to apply the data and forecast.

This model assumes the following:

  • Unlimited inoculum of Leveillula taurica is present in the field.
  • The tomato cultivar in use is susceptible to Leveillula taurica.
  • There was no disease risk before the first day of data to be evaluated by TomatoPM.
  • Fungicide sprays can reduce infection in progress and provide effective protection against the disease for 10 days.
  • Location of sensors in the field and sensor measurements are representative of tomato canopy microclimate in the rest of the field.

Effect of sprays on the forecast

When you spray a fungicide, enter the date on which the spray is made. The model assumes that fungicide sprays can reduce infection in progress and provide effective protection for a 10-day period, where the first day is the date of the spray.

As you update the weather data and the forecast, the display will always inform you of the conditions for each day and the disease risk without fungicide protection. However, the forecast will also indicate whether the crop is protected by a recent spray.

Bugs in this disease program

To make it available for the 1997 season, a few shortcuts have been taken in preparing TomatoPM. These will be corrected through better error handling in a future version.

The following conditions produce known errors:

  • Weather Data file is open when you're trying to Add Data and Update Forecast.
  • The Weather Data file you originally specified is not available in the directory (folder) under the name originally given.

Avoidance is the easiest way to solve these errors: close your data file and don't move it after you've designated it as a TomatoPM location. However, if you do get the error, the most straightforward way to fix either of these errors is to close TomatoPM without saving changes, fix the problem with the Weather Data file, and restart TomatoPM.

As other problems are known, information about them will be posted here.

Questions about TomatoPM? Please send e-mail to webmaster and we'll do our best to help.


| Home | Help Desk | Search | PestCast |

| About | Manage Pests | Pesticides | Resources | Funded Projects | What's New? |

UC IPM Home Page All contents copyright © 2001 The Regents of the University of California. All rights reserved.
For noncommercial purposes only, any Web site may link directly to this page or photo file. FOR ALL OTHER USES or more information, read Legal Notices. Unfortunately, we cannot provide individual solutions to specific pest problems. See Manage Pests, or in the U.S., contact your local Cooperative Extension office for assistance. Statewide IPM Program, University of California. DISEASE/DATABASE/TomatoPMdoc.html revised: March 6, 2001. Contact webmaster.