Predicting air temperature for frost warning using Artificial Neural Networks
MetadataShow full item record
One of the most important factors in crop growth is weather, therefore, weather forecasting is vital for agricultural production decision making. For crops such as peaches and blueberries, low temperatures during the bloom period can result in crop damage. Thus frost forecasts are important to provide a warning to farmers, who can then take appropriate actions to minimize damage to their crop. However there are no local short-term frost forecasting systems available at the moment. The complex and non-linear nature of the relationships between various meteorological factors cannot be easily modeled. The goal of this research was to develop a decision support system using Artificial Neural Networks (ANNs) to forecast temperatures in hourly increments from one to twelve hours for any location in south Georgia region, for which, current weather data was available.