AI Weather Forecasting for Smart Farms

Researchers at Australia's James Cook University (JCU) have developed a model to help farmers using smart devices choose the most accurate rainfall forecast for a given day to determine whether to irrigate.

Their hybrid climate learning model (HCLM) involves a probability-based network evaluating multiple forecasts for different rainfall patterns and a deep learning neural network generating a better prediction for the next day by reprocessing those forecasts.

The hybrid system was trained on 123,640 items of data from two years of forecast and weather data from Australia's Bureau of Meteorology (BOM) for 10 sites across six major climate zones.

The HCLM outperformed the BOM's climate models and three other experimental systems in terms of forecast accuracy.

Said JCU's Eric Wang, "We believe this model is the first to bring together the climate models, a probability network, and a deep learning neural network."

From James Cook University (Australia)
View Full Article

Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA