How Can Climate Modelling Create a Sustainable, Clean Earth?

Climate modelling sits at the heart of sustainability strategy, risk planning and climate change adaptation. 

The modelling turns physics, chemistry and observations into simulations that can test ideas and explore futures. 

When carried out accurately, it can help governments, cities and companies make resilient, science-based decisions about investment, adaptation and net zero.

What is climate modelling?

According to the National Oceanic and Atmospheric Administration (NOAA), models help us work through complex systems and test theories by representing the world as thousands of three-dimensional grid cells governed by equations for energy, fluids and chemistry. 

The administration states that scientists “run” these models by setting climate forcing such as greenhouse gas levels, then powerful computers solve the equations in each cell and pass results to neighbouring cells across many time steps to represent the passage of time. 

The higher spatial resolution uses smaller grid cells and higher temporal resolution uses shorter time steps, both of which demand more computing power but yield finer local detail.

According to the National Centre for Atmospheric Science (NCAS), climate models simulate the Earth system, atmosphere, ocean, land and ice, to calculate properties such as temperature, pressure, wind and humidity over time. 

“Climate models are based on physical principles that have been understood for many years, such as the conservation of energy,” says NCAS.

“These principles are the foundation of our understanding of climate and weather.

“Climate models are also checked by running simulations of past events. 

“We have real-world observations from the past, so we can test if the climate model is accurate or not.

“Lastly, climate models have successfully predicted patterns in our climate, such as the El Niño phenomenon.”

The NOAA’s modern Earth System Models extend this by including biogeochemistry to capture carbon cycle and aerosol feedbacks more fully.

According to the Intergovernmental Panel on Climate Change, scenarios are used to explore possible futures by varying socioeconomic pathways and the resulting greenhouse gas concentrations. 

How to use climate modelling

Once a model reproduces past climate through hind-casting it can be used to project future conditions under chosen scenarios, providing probabilistic statements about the likelihood of being warmer or cooler and wetter or drier in given regions, according to NOAA.

According to NCAS, this lets scientists test understanding, quantify uncertainty and explore the impacts of different warming levels, for example assessing outcomes at 1.5°C above pre-industrial temperatures. 

It is said that the results are best interpreted as ranges across multiple models and experiments, since structural differences between models produce a spread yet consistent trend.

According to the UK Met Office, climate models solve complex mathematical equations and cannot represent every real-world detail, so approximations are used and continually improved as science advances and computing power increases.

The Met Office also suggests that improvements in resolution now allow regional detail down to tens of kilometres and much longer runs that reproduce the last 150 years and project the next century or more. 

“We looked at projected future changes in climatic drivers for floods, droughts and wildfires across many regions of the world and found no dependence on climate sensitivity in the majority of cases studied,” says Dr Jeremy Walton, lead computational scientist for the UK Earth System Model (UKESM), on a climate modelling project he helped conduct.

“We deduce that using models with a high climate sensitivity for regional climate projections is just as valid as using those with a lower value for this property.”

Companies using climate modelling

McKinsey’s Planetrics tailors scenario analysis to an institution’s exposures and operations, combining software with sector expertise and coverage of more than 90,000 securities,  over 300 sector models an more than 250,000 emission factors.

NASA’s Goddard Institute for Space Studies notes its coupled atmosphere–ocean models feed IPCC work and CMIP archives. 

PwC describes climate analytics and scenario testing, powered by collaborator Jupiter Intelligence, to quantify physical and transition risks and support ESG reporting. 

IBM has also collaborated with NASA on an AI foundation model for a variety of weather and climate use cases, contributing to the Oak Ridge National Laboratory, the model offers a flexible, scalable way to address a variety of challenges related to short-term weather as well as long-term climate projection.

"This space has seen the emergence of large AI models that focus on a fixed dataset and single use case, primarily forecasting,” says Juan Bernabe-Moreno, Director of IBM Research Europe (UK and Ireland) and IBM's Accelerated Discovery Lead for Climate and Sustainability. 

“We have designed our weather and climate foundation model to go beyond such limitations so that it can be tuned to a variety of inputs and uses.

"For example, the model can run both on the entire earth as well as in a local context. 

“With such flexibility on the technology side, this model is well-suited to help us understand meteorological phenomena such as hurricanes or atmospheric rivers, reason about future potential climate risks by increasing the resolution of climate models and finally inform our understanding of imminent severe weather events."

Siemens highlights the use of digital twins and climate modelling to guide the Siemensstadt Square district design via Siemens Xcelerator. 

AWS outlines a Customer Carbon Footprint Tool to measure, forecast and report emissions from cloud use. 

Google notes that energy demand modelling supports its 24/7 carbon-free goal through smarter procurement and grid integration. 

BlackRock states that Aladdin Climate integrates climate science with financial risk modelling to assess portfolio exposure.