With the increasing interest and need for accurate long-term air temperature forecasts, scientists and meteorologists are finessing their methodologies to provide better and more reliable predictions. A ground-breaking approach has demonstrated enhanced forecasts for one to five weeks in advance. What’s surprising is that this leap forward is attained not by using new model simulations, but by tweaking conventional models in a more competent manner.
This latest development in climatology pertains to a process colloquially referred to as “model tuning”. Climate models are designed to simulate real-world conditions, but they necessarily contain uncertainties. Thus, model tuning is used to adjust parameters within the model to make output match observations more closely. However, current tuning methods don’t consider long-term forecasting, which leaves a significant distance between potential and applied capabilities.
In efforts to bridge that gap, researchers have unveiled a methodology that allows long-term air temperature predictions, without the need to develop new model simulations. This, researchers hope, can make substantial strides in more accurate long-term weather and climate forecasts, that are critical for ecosystems and economies worldwide.
The research, published in Geophysical Research Letters, introduced a novel tuning concept dubbed ‘envelopment’, put forth by a team of scientists led by Dr. Lukas Brunner of the University of Exeter. The envelopment process works by using a climatology envelope of air temperature probabilities instead of a single deterministic value. This methodology allows every weather forecast to fall within a particular range of possibilities, thereby enhancing its accuracy.
Notably, this particular study is concerned with air temperature projections, a key factor affecting global climate change, and one which markedly influences various aspects of everyday life, farming, health and the economy. Dr. Brunner says, “Predicting air temperature—even at local scales—provides all sectors of society with critical environmental intelligence.”
This significant shift from single deterministic values to a broader envelope of possibilities is the key to refining future temperature predictions. The team worked with the general circulation models and used recent past data to create a pool of likely outcomes which were then put through a series of tests to check their validity and reliability.
The concept of envelopment is reminiscent of an idea put forth by renowned physicist Niels Bohr, who famously said, “Prediction is hard, especially about the future.” With this new technique, meteorologists are not predicting a definite outcome but are instead offering a range of possibilities, thus improving the chances of their forecast being accurate.
Testing their innovative approach on the European Centre for Medium-Range Weather Forecasts system, the team found their results were as good, if not better, than the Center’s predictions that use a different model. Furthermore, their work reasonably predicted the historic heatwave events of 2003 and 2010, which were initially missed entirely by existing deterministic models.
This development has major implications for a broad array of sectors, including agriculture, energy, tourism, and insurance, among others. For instance, more accurate, long-term weather forecasts can help farmers plan their activities, allow energy companies to manage their stocks and operations better, and enable insurers to assess risk more precisely.
“When we’re talking about a period of one to five weeks in advance, we’re discussing the subseasonal-to-seasonal timescale—where accurate forecasts are most challenging to achieve, yet most crucial,” says Dr. Brunner.
While work remains to be done, the results of this innovative study suggest that the approach of envelopment holds significant potential for improving the accuracy of long-term air temperature forecasts without reinventing the wheel of model simulations. This breakthrough could usher in a new era of meteorology, making long-term and robust forecasting a norm rather than an exception in weather prediction services. With better forecasts, we can be better prepared for what lies ahead in our ever-changing climate.
Original Source: https://phys.org/news/2026-04-air-temperature-weeks-advance-simulations.html
