Quantum Computing Takes on Flood Predictions in the UK
A new partnership between leading quantum company Multiverse Computing, Moody’s Analytics, and Oxford Quantum Circuits (OQC) has secured funding from Innovate UK to leverage quantum computing techniques in the development of advanced flood prediction models, offering a solution that surpasses the constraints of traditional methods.
Overseen by the UK Department of Environment, Food, and Rural Affairs, this initiative aspires to boost the UK’s resilience against extreme weather phenomena intensified by climate change. The nation hopes to enhance its adaptive measures by employing quantum-assisted computational fluid dynamics.
The collaboration secured their position in Phase 1 of the UK government’s Quantum Catalyst Fund with their proposal titled “Quantum-Assisted Flood Modeling: Pioneering Large-Scale Analysis for Enhanced Risk Assessment.” The main objective is to harness quantum computing to tackle the computational challenges present in large-scale flood modeling. This promises a more accurate and efficient risk assessment and management process.
Leading the charge, Multiverse Computing will handle the technical aspects, including formulating the problem and developing algorithms. OQC is set to provide quantum hardware and additional resources. Moody’s Analytics, renowned for its expertise in global risk management, will supply industry knowledge, data stipulations, and valuable insights into computational efficiency.
Rising Tides with Climate Change
Experts predict that with the increase in global warming, more extreme weather events, including flooding, will become more common. Without accurate flood predictions, governments and citizens cannot properly prepare for future flooding events. This makes it difficult for towns and cities to recover better after a catastrophic flood hits. “An increased risk of flooding is one of the most challenging climate change adaptations facing the UK,” elaborated Victor Gaspar, the Chief Sales Officer at Multiverse Computing. “This risk is significant enough to have an impact on the country’s gross domestic product over time.”
Improving Flood Predictions
The current landscape of flood modeling heavily relies on two-dimensional hydrodynamical models. Based on the Shallow Water Equations (SWE), these models are crucial for predicting events like dam breaches, storm surges, and river flood waves.
However, the extensive computational demands of these simulations often limit their scalability and resolution. “Fluid dynamics problems are computationally hard to model using classical computers because of the high number of variables involved,” explained Gaspar. “Our solution will be able to manage those variables more successfully than classical solutions and offer more precise risk predictions.”
While solutions like parallel and GPU-based computing have been implemented to expedite the process, the introduction of quantum computing brings forth new horizons.
The project produced by these three collaborators will employ a Quantum Physics-Informed Neural Network (QPINN) algorithm, which fuses classical data processing with quantum processing using a Variational Quantum Circuit (VQC).
Phase 1, lasting three months, concludes on Nov. 30, 2023. Following this, Phase 2 will span 15 months, commencing in January 2024. The progression to the second phase is contingent upon the successful culmination of the first.
Gaspar added: “We expect in the future our work to inform policy makers in business and government and to support new incentives to plan differently for increased flood risk. This could include new protections for buildings on the shoreline, informed decision making about new construction in high-risk areas, and a better understanding of the risk level for properties that are not currently considered to be in flood-prone places.”
Kenna Hughes-Castleberry is a staff writer at Inside Quantum Technology and the Science Communicator at JILA (a partnership between the University of Colorado Boulder and NIST). Her writing beats include deep tech, quantum computing, and AI. Her work has been featured in Scientific American, Discover Magazine, Ars Technica, and more.