Simplifying IPM with Remote Weather Stations
Jon M. Clements, UMass Extension Tree Fruit Specialist, UMass Cold Spring Orchard, Belchertown, MA
While modern business has managed largely to insulate itself from the impacts of weather, agriculture remains an exception, very much dependent on what nature delivers as rain, sun, heat, cold, or storms. Weather not only directly impacts crop production, but also drives the development of insects and diseases that attack them. Over the past 25 years, and as part of an overall IPM (Integrated Pest Management) strategy, growers have been given the tools to manage risks related to weather and pests by closely monitoring weather information and using it to predict when, for example, pesticides should be sprayed or plants should be protected from frost with irrigation. Thus, monitoring key weather parameters, such as rainfall, wetting periods, temperature, and evaporation, can provide agricultural businesses with critical information.
Using weather information to help make pest management decisions involves collecting accurate data and analyzing it to help make key management decisions. Analysis is done using models, which are mathematical calculations and logical formulations that use weather data and other information. The model will make a recommendation as to whether some action should be taken. For example, apple growers can determine whether they need to treat for fire blight using a model. They collect daily temperatures, dew, and rainfall, and as trees come into bloom, combine this to determine whether a streptomycin spray is needed. Fire blight can kill trees and cause serious damage, so streptomycin should be used if there is risk of infection. On the other hand, unnecessary streptomycin applications waste money and increase the chances that the fire blight pathogen will become resistant to the antibiotic. Using a model such as MARYBLYT enables growers to know when treatment is needed and when it isn’t. Models have been developed for many important disease and insect problems, as well as for frost prediction, timing harvest and irrigation, or optimizing fertilizer applications.
In pest management, reducing pesticides often depends on accurate weather information. It is a cornerstone of IPM. It is equally important for organic producers. Growers using IPM or organic methods often receive a premium for their products. Increasingly European and other foreign markets are demanding food that is grown with minimal use of pesticides. Readily available weather and decision support can therefore reduce costs by decreasing inputs for pest management, and improve returns and markets for agricultural products.
Growers, however, often do not use these valuable decision support tools. In spite of technological innovations, growers find it difficult to regularly get the weather data needed and most importantly apply it in a decision support model. Most growers do not want to be bothered with the intricacies of downloading data and running models. To be useful, weather instruments have to be accurate and function day in and day out. Data has to get from the instruments to the grower or a computer in a form that can be readily used and understood. Maintaining a weather station and accessing data adds tasks to an already full day of work. Sometimes, it involves technological expertise. Most growers, when asked, say that they would and do use model-based forecasts when those forecasts come as simple recommended actions. But they find the process of gathering and analyzing the information themselves too difficult and/or time-consuming to do on a regular basis.
As a result, newsletters and other grower information sources will publish information based on weather data from selected locations or for a region. Subscription services, such as Skybit, also have arisen, and these provide growers with farm-specific information interpolated from off-site National Weather Service weather stations.
Such information is generally useful, but variability in conditions from farm-to-farm, or delays in delivering it, can make recommendations less accurate. To fully utilize the power of weather-based models, growers need a system that gathers accurate data and processes it for their particular farm, delivering easily understood recommendations in a timely manner.
Recent developments in wireless technology, robust weather monitoring hardware, and web-based software have made it possible to do this. Weather data can be obtained on a farm, sent frequently to a central computer server, where it can be processed in forecasting models, and then presented as recommendations in the form of web pages accessible by each grower.
At the University of Massachusetts, we are currently evaluating two web-based weather stations from Onset, one which is now located here at the UMass Cold Spring Orchard Research and Education Center in Belchertown, Massachusetts, and the other at Tougas Family Farm in Northboro, Massachusetts. Both systems are currently online and transmitting weather data to the Internet via integrated Wi-Fi communication modules. Live data from our both Onset stations can be viewed at:
As mentioned earlier, most growers do not want to deal with the complexities of retrieving data and running models. Rather, they want to simply go to a web page in the morning, find out what’s happening in the field based on the environmental conditions, and take action. The remote systems we have recently deployed make this possible, and are good examples of the types of technology tools that growers can take advantage of today. We are currently soliciting funding to develop application(s) that collect current weather data provided by the Onset stations, run insect and disease models, and then output the results to user-friendly and immediately accessible web page(s).
From a research perspective, having easy access to information from different field sites is also very valuable. We are able to get a real-time view of field conditions, and at the same time are able to collect long-term data that can be used for trend analysis and correlating insect and disease incidents with on-site weather information. Thus, we can use this information to fine-tune current models and develop new ones.