- Participants at COP27 should recognize the importance of collaborating to improve technology essential in confronting climate change.
- Humanity must invent its way out of the climate problem — but this is only feasible if it is done collaboratively, and with a focus on cutting-edge technology.
- A true climate network — a multi-nation, multi-disciplinary partnership — can shape the technology and action needed to end climate change.
Global summits like COP27 are crucial to getting the world on the same page, driving progress toward climate change mitigation and adaptation.
This year, there will be many discussions about funding energy transition, assessing progress on the nationally determined emission limits and transferring money between nations for climate-related projects.
This is good — but it is not enough. It is crucial that science and technology are central to the discussion. Humanity must invent its way out of the climate problem.
Have you read?
Inventing our way out of climate change
That’s where artificial intelligence (AI), hybrid cloud computing and soon quantum computing can help. A serious opportunity for mitigating climate change lies in the discovery of new materials and reinvention of existing industrial processes.
One example is to capture carbon dioxide more efficiently from point sources or to introduce carbon-negative processes of making cement, steel and other products. Localized geospatial modelling and decision support systems can improve climate change adaptation, resilience and emergency response to save lives and create resilience infrastructure. Progress in these areas requires a new way of accelerating scientific discovery, and, more importantly, making digital tools accessible for innovation, anywhere.
Traditionally, we either stumble upon a new material in a lucky accident — as was the case with graphene — or spend time and money on a trial-and-error process.
For the latter, scientists rely on their knowledge, experiments and published literature to design a molecule, hoping it has the desired properties. They then follow many cycles of synthesis and testing. But the possibilities for molecular configurations are vast — more than there are atoms in the universe. That propels the number of potential new materials to infinity. The ever-surging amount of data is also vast, with some half a million new papers annually published just in the field of material science — impossible for a human to go through in a reasonable amount of time.
Mitigation and adaptation both need tech
Consider climate change mitigation. To find a new material for better CO2-filtering membranes, IBM researchers turned to AI combined with physics domain knowledge.
First, they outlined the desired properties of the membrane they wanted to get: permeability, chemical selectivity for specific gases and durability. Next, another AI model combed through the past knowledge of membrane materials, all relevant existing patents and publications. They used predictive, so-called generative models to create possible molecules that would make the material more efficient at filtering CO2. The next step was to simulate this new molecule on a high-performance computer, to confirm that it performed as expected. In the future, a quantum computer could make these molecular simulations even more efficient.
Once the design is confirmed, the final step is to create the actual membrane with an AI-driven lab, such as IBM RoboRXN. It combines AI, cloud computing and robotics to help researchers design and synthesize new molecules anywhere and at any time.
But mitigation alone is not enough. Some impacts of our changing climate are here to stay — we need to learn to live with them.
Partnerships are already working
When it comes to climate change adaptation, there is one key ingredient to drive progress: partnerships, to widely deploy data and AI technologies.
IBM is working with several partners on innovation, including the UK’s Science and Technology Facilities Council (STCF), Environment and Climate Change Canada (ECCC) and African Risk Capacity (ARC). They are pursuing faster decarbonization approaches and driving actions related to adaptation and resiliency together.
But most climate applications need data from multiple storages and places around the globe. These are government agencies, national labs, Earth imaging companies and research institutions, all generating petabytes of climate-relevant data every day. AI can help make sense of this distributed data through collaboration — call it a ‘global climate network.’ IBM researchers and partners are now at the early stages of exploring the technical feasibility of this idea, with the ongoing partnerships.
The STFC-IBM team is leveraging innovations in indexing multidimensional climate data to rapidly discover climate-relevant information such as from aerial imagery, maps, IoT, drones, LiDAR, satellites, weather predictions and climate change projections. The tech is poised to accelerate the discovery of such data by a thousand times, advancing our understanding of climate change impact and saving lives.
Take flooding. In the UK, some 1.8 million people live in areas at high risk of deadly floods, which cause about £1.6 billion worth of damages annually. Building on previous work on flood detection that uses the climate data discovery tech tool, the team used a new AI model to track UK flood events in near-real time. The researchers used data from UK floods mapped from the semi-automated European Union Copernicus Emergency Management Service, including the flood that devastated York in 2015. The AI modeled the data with an average accuracy of 98% compared to ground truth.
IBM Research and ECCC are using AI to compare historical weather data, accelerate understanding of severe weather and to better forecast future extreme events.
And IBM is also starting to collaborate with the African Risk Capacity (ARC), building on the IBM strategic partnership with WITS University to improve climate risk modelling and predictions on the African continent. Africa has been particularly hard hit by climate change, with most countries lacking an early weather warning system, so better modeling and prediction with AI will be important tools in the future.
Science and scientific talent know no borders and the technologies being developed could be accessible everywhere globally.
To accelerate climate research and applications further, we must come together to shape the vision and technical feasibility of launching a true climate network — a multi-nation, multi-disciplinary partnership. Only together will we manage to bring our sustainable future closer, faster.