Potential frost damage during the growing season poses a serious threat to cranberry crops in Wisconsin. Cranberry growers require advance warning of impending overnight frost conditions so that frost protection activities can be properly coordinated. For this purpose, we have developed an automated system that delivers timely frost forecasts to growers via the World Wide Web. An initial forecast, generated by the ALEX soil-plant-atmosphere model (adapted to a cranberry bog microclimate), is disseminated over the Web daily at around noon. This initial forecast is then updated hourly throughout the rest of the day as new surface and satellite observations become available; surface observations are used to update upper boundary conditions to the ALEX model, while GOES satellite images provide improved radiation inputs. The system also incorporates a feedback loop for correcting persistent model biases. Each day, archived forecasts are compared with surface observations obtained during the previous month; new forecasts are compensated for significant detected biases. This system has demonstrated good skill in predicting overnight frost conditions in two test bogs in Wisconsin, with significant improvements in accuracy gained through the assimilation of real-time weather data.
Special Session -- Weather Data Requirements for Integrated Pest Management