This article is more than 1 year old

BASF looks to quantum computing to improve weather modeling

Chemical giant's work with Pasqal could help improve climate change models too

A new quantum computing partnership could pave the way to more efficient climate change modeling. 

BASF and French quantum computing startup Pasqal have linked arms in a move that will see the chemical giant use Pasqal's proprietary quantum tech to improve its ability to predict the weather.

BASF's chemicals are found in a variety of products and materials. One of its sectors involves a portfolio of farming optimization software that includes xarvio Field Manager, a crop optimization platform. BASF uses weather models to inform its software's crop growth simulations, pesticide drift and other metrics, and believes Pasqal's quantum systems could help.

Physics-based weather modeling is complicated. Wind data, heat transfer, solar radiation, humidity, topology, and other factors combine into complex nonlinear differential equations that need solving, Pasqal said. 

"Pasqal aims to solve the [math] in a novel and more efficient way by implementing so-called quantum neural networks on its neutral atom quantum processors," the company said. 

There's already a non-quantum model that does what Pasqal and BASF are aiming toward: Nvidia's Earth-2 supercomputer. As opposed to using quantum algorithms, Nvidia's climate prediction model relies on physics-informed neural networks, which are commonly used for weather and climate modeling.

Dr John Manobianco, senior weather modeler at BASF's agricultural solutions division, said BASF's work with Pasqal would help it simplify computational simulations "once quantum hardware matures to a point where we can actually leverage these algorithms." 

A climate change modeling solution?

Pasqal said the information it learns from the collaboration with BASF can "build a foundation for future extensions of Pasqal's methods to support climate modeling." 

Benno Broer, chief commercial officer at Pasqal, told The Register that Pasqal's work with BASF will "parameterize, implement and test Pasqal's proprietary family of quantum algorithms for solving differential equations." 

Pasqal's differential equation solving algorithm has also been used by BMW to reduce vehicle component testing time. Along with its partnerships with BASF and BMW, Pasqal also announced an agreement with Saudi Aramco in March, and claims Johnson & Johnson, LG, Airbus as other customers. 

With BASF, Pascal's differentiable quantum circuit (DQC) algorithm will be turned to predicting weather patterns, which Broer said will lead to Pasqal being able to specialize its algorithm.

"We expect to be able to tailor [DQC algorithms] to model such patterns on multiple spatial and temporal scales, meaning both long-term and short-term, local and global patterns (including climate change patterns) and events," Broer said. 

With recent climate events bad enough to knock Google and Oracle data centers offline in Britian and do lasting damage, faster and longer-term weather modeling is definitely needed. ®

More about


Send us news

Other stories you might like