• Last day for blocking high. Is Ai coming to our weather forecasts?

    Last day for blocking high. Is Ai coming to our weather forecasts?

    The welcome sunshine on Wednesday morning lasted until late afternoon, through times of variable cloud, that pushed the thermometer to reach a high of 17.0C at 14.51, which was again well above my long-term average at + 6.3C. It was the 22nd dry day this month with the UV level again rising at its peak up to the ‘Moderate’ category.

    Overnight saw the thermometer slowly slide downwards to reach a low of 1.9C at 04.04. Around that time, looking at the cloud radar from that time, thin cloud and then thicker cloud drifted across from the east, as a result the temperature began to slowly recover and the humidity increase so that by 08.00 the thermometer read 8.1C.

    Thursday revealed a totally cloudy sky with calm conditions.

    The high pressure is stable at the moment, recently acting as a blocking high, but as the afternoon progresses, and overnight, it will slowly fall as it gives way to a depression, that by this time tomorrow will have developed just to the north of Scotland. The low pressure system will throw a couple of cold fronts down across the country that are likely to produce the first rain since the 23rd and cooler conditions. We will see by tomorrow if the additional rainfall will push the 2025 March rainfall total above the 2011 record low?

    I recently came across the following news item. We will see if the future provides a more accurate weather forecast.

    Aardvark Weather, a new Ai model developed by researchers in the UK and Canada, could mark a turning point in global weather forecasting by replacing traditional weather simulations with artificial intelligence to maximise cost efficiency and accuracy.

    Unlike conventional forecasting tools that simulate atmospheric physics through complex equations, Aardvark Weather is a ‘deep learning’ model that generates global forecasts for wind, humidity and temperature at multiple pressure levels.

    They state “By making these components more efficient, Aardvark could run forecasts more often and at higher resolutions, improving speed and accuracy”.

    They comment that “Aardvark can generate a full global forecast in just one second compared to the many hours needed by older models, which would result in a drastic reduction in computing requirements”.

PREVIOUSLY