Using Forecasting Models to Time Tomato Fungicides (and Insect Management)

By Jude Boucher, UConn Extension Educator, Commercial Vegetable Crops

In the last issue of ‘Crop Talk’ I mentioned that “there is a computer model (Tom-Cast), based on temperature and leaf wetness time, that can help predict early blight (EB) infection periods”, as well as Septoria leaf spot (SLS) and Anthracnose ripe rot (ARR) on tomatoes. There is actually a second blight cast model (Simcast) that can help time sprays for late blight (LB). I also said that I wasn’t a big fan of these models for predicting when to apply fungicides because scouting has proven itself to be more reliable over time and capable of alerting you to the presence of pests that the models can’t. However, I tend to be a low-tech type of guy…I don’t even carry a cell phone! Some of you high-tech folks may actually want to monitor disease development with computer models, and thanks to Mary Concklin (UConn Fruit IPM) and the folks at Cornell University, you are in luck.

Cornell has set up the Network for Environment and Weather Applications (NEWA), which provides the weather data from nearby farms throughout the Northeast to all network users and runs the models for you. Mary and other members of UConn’s IPM team received a large grant from USDA EIPM, that allowed her to set up a network of 41 weather stations here in CT, located at farms and schools, that are now (or soon will be) linked to the NEWA system (at least for 2014). This means that there is now local weather data from locations near you, and the model predictions from nearby sites should apply to your farm. How much easier could it be?!!! You don’t even have to buy and run a weather station, deal with all the electronic headaches, or pay to link into NEWA (UConn IPM pays one price for the state to participate). All you need is an online computer and you can get the information for free!

Another nice feature of this system is that it will run degree-day insect development models for certain pests, such as cabbage and onion maggots, which can help you determine when each generation of the pest should occur this year, thus allowing you to more accurately time your scouting and control procedures. I should mention that the Tom-Cast and Simcast models also work for potatoes, because they too have problems with early and late blight. The site can also help you with on-ion diseases, downy mildew on cucurbits, and Stewart’s wilt of corn.

I could just tell you that all the information about how all these different models work, and how to interpret them to help time your control procedures, is available on the NEWA site (, but I’ll give you a little summary of the Tom-Cast, Late blight and degree day models to entice you into looking at the site (in case you’re a low-tech type of person like me).

Tom-Cast for timing fungicide applications for EB, SLS and ARR: This model starts with a caution that it should not be used if you have a bacterial disease (canker, spot or spec) in your plantings because the fungicide spray intervals recommended will be too long for copper (the best bactericide). As mentioned, temperature and leaf wetness duration are used to calculate disease severity values (DSV’s), and when you accumulate enough DSV’s at your site you make a fungicide application. For example, it might take only 3-5 hours of leaf wetness at the near-optimum temperatures for disease infection of 69-77°F to produce one DSV, while it takes 4-8 hours between 64-68°F or 78-85°F, and 7-15 hours between 56-63°F to produce 1 DSV. You can also accumulate 2, 3 or 4 DSV’s in the 69-77°F range, or any other range, with longer wetting periods up to 23 hours. But you don’t have to worry about that part. The machine calculates the DSV’s and keeps a running count by date. You spray for the first time when it accumulates 25 DSV’s starting at transplanting, and then every 20 DSV’s after that if you’re using Bravo, Quadris or Cabrio, and every 15 DSV’s if you are using a mancozeb-type product (i.e. Dithane or Manzate). Make an application just before the next rain if the spray interval is longer than 14 days. Also, make the first application by July 11th if you haven’t accumulated 25 DSV’s by then.

To use the Tom-Cast model, start by going to the tomato disease page and read a little about the model. Then, at the top, select “Pest Forecast”, then on the drop down menu choose “tomato diseases/Tom-Cast”. Fill in the date you are interested in and select the CT site nearest to you from the drop down list. It will provide you with several columns of data, but the one you are interested in is the accumulated DSV’s. Note: there is nothing accumulated for 2014 yet, so you will have to choose i.e. June, 2013, Middlefield, CT (where they had a station running last year) to see a page of site specific data.

Simcast for timing fungicide applications for late blight: The LB model is slightly more complicated using both “Blight Units” to indicate disease severity and “Fungicide Units” to indicate when your fungicide protection runs out. It assumes you’re using a chlorothalonil-type product, such as Bravo, unless you enter a different product. It also takes into consideration how susceptible, tolerant or resistant your tomato variety is to LB.

To use Simcast, choose the Simcast model instead of Tom-Cast, then, enter the weather station site nearest you, the date of your last fungicide application, and your tomato variety from the drop down list of choices. If your variety is not on the list, or you have a mix of varieties, assume it is, or they are susceptible, and enter that. Then hit the “get report” button and you will see accumulated Blight Units and Fungicide Units by date. The units (numbers) will turn red when the spray threshold is reached.

Degree-day models for insects: Degree-day models are a way to predict pest or plant development based on temperatures. In its simplest form, the degree-day calculation uses the average daily temperature (maximum + minimum/2) minus the pest’s minimum development temperature (base temp). So, cabbage maggot begins to resume growth or development at 4°C and onion maggot at 40°F. For example, for cabbage maggot, a day in March might have a high of 8°C and a low of 2°C (8+2/2 =5) minus base 4°C, would result in just 1 DD accumulating on that date. The NEWA system actually uses a more accurate degree-day model based on algorithms, but again, you don’t have to worry about calculating degree-days, because the NEWA system will do it for you, and just show you how many have accumulated this year or since the last generation of the insect.

So for cabbage maggot, the overwintering generation will start to produce the first adult flies after 161DD. The first generation flight will peak at 251DD and end at 449DD. You want to apply the first application of cabbage maggot fly bait (i.e. Seduce) just after 161DD, or a drench before peak flight because the eggs will begin to hatch by then. The second, third and fourth generation flights will end 508, 465 and 399DD after the proceeding generation. The NEWA system recommends you reset the accumulating degree-days to zero at the end of each generation’s flight to help keep track of the next generation. Usually we only have to treat during the first flight because soil temperatures later in the season will normally be above 90°F, which will kill the eggs and hatching maggots (unless prolonged rainy periods suppress soil temperatures while transplanting later plantings).

To use the degree-day models, just choose “pest forecast” at the top of the page, and “cabbage maggot” from the drop down menu. Then choose the closest weather station site from the drop down menu and it will provide the accumulated DD by date. At this time, you can’t access the data for the previous year (i.e. 2013) for the vegetable insects, as you can for some of the fruit insects, so you can only compare this year to last year for some insect pests. However, Cornell will be working to improve the site in the next year or two, so you should see improvements as you use the system.