DEPARTMENT FOR SCIENCE, INNOVATION AND TECHNOLOGY

AI for Climate - Calls

Project disclaimer

Disclaimer: The data for this page has been produced from IATI data published by DEPARTMENT FOR SCIENCE, INNOVATION AND TECHNOLOGY. Please contact them (Show Email Address) if you have any questions about their data.

Description

New multi-year activity ( this proposal covers FY24/25 only) using AI to advance weather and climate science, and hence deliver improved predictions and projections more quickly than would otherwise be possible. AI and data science is of considerable interest to several developing countries, both in terms of developing their expertise in this area and for the potential benefits it offers in terms of reducing the need for expensive high performance computer resource. The activity aims to work with partner countries ( India [ODA] & South Africa) to co-develop region-specific climate models with AI/ML built in to equip in-country partners with the capability to better understand climate risks. This work will also result in models that can be used by other ODA-eligible countries to support their current capability to understand and adapt to climate change.

Objectives

In recent years, there has been a transformative improvement in the capabilities of Artificial Intelligence (AI) and Machine Learning (ML) which offer potential in many fields. Over the last 18 months we have seen an explosion in the application of data driven techniques to the weather prediction problem, with considerable success in a short period of time. There is similar scope to provide a step change in climate change projections with the potential to provide more local detail, more accurate projections and explore a wider range of future scenarios. This large multi-year project is proposed to significantly advance the science in this area and deliver tangible outputs in line with the aims of ISPF. A number of strands (work packages) are proposed which explore several promising options for the use of AI to advance climate science, and hence deliver improved projections more quickly than would otherwise be possible. For the UK, advancement in this space will enable a more complete, accurate and rapid response to questions posed by policy makers, industry and the public around climate change. It will also continue to cement the UK’s position as a world-leader in climate modelling. For developing countries around the world, it will provide the opportunity to produce local climate change projections quickly and easily, not requiring expensive high performance compute resource. For the partner countries and institutes, it will also enable the sharing of data science knowledge and skills and the undertaking of collaborative, cutting edge, scientific research. key ODA Outputs: 1) ML downscaling model to provide local climate projections for anywhere in the world. 2) ML climate model capable of producing climate change projections given input forcings. Both activities will significantly help developing countries. The downscaling activity will enable high resolution predictions for other parts of the world and Africa is likely to be one region used for training. As such, we anticipate South Africa being a key partner country, building on existing collaborations which share the UK’s traditional (physical) weather and climate models. Like AI4NWP (the use of AI for numerical weather prediction and weather modelling), the foundational climate model (once trained) is likely to allow partners in developing countries to run their own climate simulations without requiring a large amount of compute resource.

---

Location

The country, countries or regions that benefit from this Programme.

Developing countries, unspecified

Disclaimer: Country borders do not necessarily reflect the UK Government's official position.