Predictive tools for informed decisions

Since 1996, Oregon IPM Center has been developing and maintaining a weather and climate driven decision support website for pest management and related agricultural needs. We now host nearly 150 predictive pest and disease models to help growers know when and where to act. 

Oregon IPM Center has been developing and maintaining a weather and climate driven decision support website for pest management and related agricultural needs for nearly 25 years. Beginning with a small number of insect models, this program has grown into a complex system of models, weather networks and advanced features. As of June 2017, there are over 140 pest and crop models (see details below) integrated with over 29,000 real-time weather stations from hundreds of weather networks across the United States. These searchable models integrate weather forecasts from several sources.

 

Online Pest & Weather Modeling Tools

Currently, we have 15 invasive insect models, 47 insect pest models, 32 crop models, 24 disease risk models, 5 weed models, 2 tree fruit dormancy (chilling requirement) models, 2 predator mite models, 1 endangered species (a butterfly) model, 3 pesticide drift model prediction aids, a soil solarization model (currently for 2 species of Phytophthora), a grass seed stem rust simulation model, and 1 mating disruption dispenser model in the system, along with generic degree-day and plant disease risk models for exploratory research needs. Weather data is obtained in real-time from 100’s of weather networks totaling over 29,000 weather stations.

    Project Goals

    • To provide a free, comprehensive website for accessing weather and climate models of all types to support IPM decision-making
    • To develop a nation-wide IPM resource that includes access for under-served regions and cropping systems
    • To develop models that accurately predict key pest development points as an aid in management decision making
    • To integrate both site-based and mapping-based tools for local and synoptic/region-wide services
    • To integrate multiple models, tools, and features in one place, including:
      • multiple models with common inputs and outputs
      • short and long-term forecasts
      • climate data
      • comparisons with recent and past predictions
      • quality assurance
      • graphical and tabular outputs
      • and full transparency of the input weather data used to drive models.

    Key Activities and Outputs

    Model Building

    Based upon requests made by growers, extension, researchers, and others, we select “best available” research and field monitoring data to build new models for the system, and independently validate the models whenever possible. The models are then readily added to the modular, open-source software system.


    Weather Database

    Most weather data is public and free for non-profit use as supplied by the Utah MesoWest system. We also custom-integrate several agricultural weather networks that are not part of Mesowest. There is no singular state-run agricultural weather network for the state of Oregon at this time, and Oregon IPM Center has no actual weather station hardware maintenance to perform. This is undertaken by participating and cooperating networks, including:

    AGRIMET (Bureau of Reclamation, US Dept. of the Interior), AgWeatherNET (Washington State University), ADCON (A private weather station hardware manufacturer), CIMIS (California Irrigation Management Information System), California PestCast (Univ. California crop disease management network), ASOS/METAR (Automated Surface Observing System, using METAR formatted data), RAWS (Remote Automated Weather Station, run by several US Federal Agencies), APRSWXNET/CWOP (Automatic Position Reporting System Weather Network/Citizen Weather Observer Program)


    “Added Value” Features 

    There are a number of innovative features such as:

    • Short and long-term forecasts, including 2 separate 7-day forecasts, a 3-month (CFSv2) extended forecast, and a 7-month (NMME) extended forecast
    • Interactive charts that show a range of projected outcomes using 7 different methods to forecast weather data
    • A system to fill in missing data with “virtual”, estimated data
    • A “Degree-Day Clock” tool that allows comparison of current season heat units to past years (e.g. reporting that we are for example “10 days behind last year and 10 days ahead of normal”)
    • Gridded map products: including interactive degree-day maps

    These features are built using funding from purpose-driven grants (largely from USDA NIFA, USDA SARE, and USDA-APHIS-PPQ), and priorities are determined through feedback from growers, researchers, extension agents, and other end-users who request models and features to support their pest management production, research or extension needs.
     

    MyPest Page

    The MyPest Page tool is programed to be able to integrate new models and features, this prototype feature provides an entry point for IPM, whereby multiple insect pests, diseases, weeds, crops, and beneficial organisms can be modeled from a single web-based user interface.
     

    Training and outreach

    This system is regularly demonstrated to growers at events around the PNW region, and several “webinar” tutorials are on the website that demonstrate how the tools can be used. Trade publications often highlight the system as it applies to one or a few models. One-on-one training via the phone remains a good way to initiate a new user to use the website to meet their particular needs.