Artificial Intelligence is taking weather forecasting by storm

Less than a year old, low resolution AI based weather models are already defeating physic based super computing models within a 7 day timeframe. 7 days ago, the highly fancied American GFS model had a deep Tasman low impacting nationwide, where the AI model from ECMWF projected it moving off the east coast. AI defeated the GFS model resoundingly. The world of weather forecasting is changing rapidly and for the better.

Here is a quick summary on what is the difference between AI and normal physics based weather models.

AI weather models and physics-based weather models differ primarily in their approach to predicting weather. Physics-based models use mathematical equations based on physical laws, such as those governing fluid dynamics and thermodynamics, to simulate atmospheric processes and forecast the weather. These models require vast amounts of data and computational power to solve complex equations. In contrast, AI weather models use machine learning techniques to identify patterns and relationships within large datasets of historical weather data. Instead of relying on explicit physical equations, AI models learn from past weather events to make predictions. While physics-based models focus on the "why" behind weather phenomena, AI models emphasize the "what" by finding correlations in data.