Taken from Agronomist & Arable Farmer, March 2017
Two new models are to be added to Growers Choice Interactive (GCI), Agrovista's online pest and disease forecasting service for top fruit growers.
The apple sawfly model had a successful test season in 2016 and will be included in the GCI package this spring.
""Apple sawfly can be a devastating pest," says Agrovista fruit agronomist Alex Radu. ""Provided pesticide applications are made at the right time you can get very good control. But if you miss it, the opposite is true.""
Eggs are laid at the end of flowering, he explains. Larvae hatch as fruits start to develop, and burrow into them, sometimes destroying whole clusters of fruit.
""Spraying needs to take place after eggs are deposited, but before they hatch. GCI collects weather data from all the key top fruit-growing areas and feeds this data into the model to help assess the optimum time for application.""
A powdery mildew model is also being developed. It will be tested during 2017 and, provided all goes to plan, will be added to the GCI package in spring 2018.
GCI supplies continuously updated information using live weather data from a network of stations, combined with the latest computer modelling techniques to ensure efficient, timely and effective use of plant protection products against a range of pests.
As well as the new models, GCI covers scab, codling moth, fireblight and canker. It also offers a range of weather data and a 10-day weather forecast.
The subscription service also includes text messaging to alert growers of major events and telephone access to a trained agronomist.
Local forecasts under test as new data source
Agrovista is investigating the performance of web-based, ultra- localised forecasts to compare the accuracy of their output with that of physical weather stations as a data source for GCI pest and disease modelling.
Agrovista fruit agronomist Tom Johnson says one weather forecast service under examination could already provide data parameters based on a 3km grid.
These include hourly temperature, rainfall, relative humidity and wind speeds-all the information currently supplied by actual weather stations, he explains.
""It is possible to apply advanced leaf wetness simulations that are key to our GCI disease models," says Mr Johnson. ""We have tested these local forecasts at 35 sites across Europe, and found temperature data to be very accurate for insect models and ascospore maturation.
""Relative humidity values were lower than the actual weather station recordings, and there were some differences in rain event recordings due to the scale of the grid. As a result, infection risk was overestimated.
""We will continue testing the ultra-local forecasts in the UK this season. With further refinement, we believe they may have the potential to provide reliable local data that will produce accurate weather forecasting and pest and disease predictions.""