Predicting the unpredictable

A better system for predicting rainfall patterns could help mitigate the consequences of climate breakdown.

The basis around which climate models have been built over the last 30 years misses some fundamental physics that we now know is essential for reliable predictions. 

Experts warn our ability to deal with climate change lies in jeopardy, without drastic action to improve rainfall predictions. 

It’s unknown to scientists what may happen to rainfall in the coming years. Severe floods, as well as prolonged droughts, are already defying expectations despite decades worth of research.

Experts argue that the answers exist but a huge joint international investment in resources, expertise, and infrastructure – amounting to an estimated $250 million annually – is urgently needed to develop advanced climate models.

Quantum leap

Lead author Professor Dame Julia Slingo, of the University of Bristol’s Cabot Institute for the Environment, said:

“The basis around which climate models have been built over the last 30 years misses some fundamental physics that we now know is essential for reliable predictions.

“The solution is within our grasp; we must take a quantum leap from our current 100-kilometre-scale global climate models to 1-kilometre-scale models.

"At these scales, the complex physics of rain-bearing systems are properly represented for the first time. The consequences reach far beyond the future of our water, to many aspects of climate change.”


The team request the creation and resourcing of a federated group of leading modelling centres, linked to dedicated, pioneering Exascale computing and data facilities.

This sophisticated climate prediction system would serve all nations, providing robust evidence across all aspects of climate change.

Professor Stephen Belcher, co-author and Chief Scientist at the UK Met Office, said:

“The scale of the task is formidable. Even though our scientific understanding has moved on, as have the technological developments in computing and data storage, this endeavour requires an international effort.”

Changes in the seasonality and natural variability of rainfall can have profound effects on many living systems. This threatens food security, water security, health, and infrastructure investments.

Yet how little we know about the future of our water was crystalized in the recent Assessment Report of the IPCC. This shows substantial uncertainties in rainfall changes, especially at regional and local scales.


Professor Thomas Stocker, co-author and a former Chair of an IPCC Assessment Report working group, said:

“The twin goals of net-zero and climate resilience require a substantial acceleration in the delivery of reliable and actionable climate information, particularly for the most vulnerable regions.

The current climate models cannot provide this, but global investment and scientific partnerships in kilometre-scale global modelling will make this a reality.”

Co-author Paul Bates, Professor of Hydrology at the University of Bristol, added:

“This proposed investment pales into insignificance against climate-related losses, even today. It represents about 0.1% of the estimated annual costs of hydrological extremes, not counting the lost lives. These costs will only rise as climate change continues to bite.”

This Author 

Yasmin Dahnoun is an assistant editor for The Ecologist. This article is based on a report by the University of Bristol Cabot Institute for the Environment. 

More from this author