Building Digital Models to Navigate the 21st Century’s Complex Ecological and Social Systems

Humanity created, captured, copied, and consumed more than 64 trillion gigabytes of data last year. This deluge of information is being used to try to model the world around us in unprecedented detail. That includes complex systems like cities, ecosystems, and the climate.

Going forward these models will become increasingly intermeshed, creating sprawling socioecological simulations that can provide policymakers with invaluable foresight on the outcomes of economic, environmental and social policies. While those simulations, often referred to as “digital twins”, can provide knowledge about the potential evolution of a system, big data and machine learning approaches have so far failed to capture the full complexity of real-world situations and different feedback loops. Finding ways to combine models with different scales and purposes and ensuring that today's biases and prejudices are not baked into them, will require a sustained interdisciplinary effort that includes full engagement among citizens.

Many initiatives for “digital twins” have been recently launched. To what extent will these initiatives be able to reproduce the complexity of real-world systems?
Can we combine models of physical reality with those simulating more intangible social phenomena?
How reliable are today's leading models and how can policy makers use them wisely?
How can we ensure models used to guide policy are transparent, equitable and explainable?

Takeaway messages

“Digital twins function as experimental landscapes that let scientists analyse risks, support decision-making and foster disaster resilience, which is becoming important to adapt to climate change. ”

“There are limitations from being obstructed by biases, randomness, turbulence, chaos theory. It will probably never be possible to produce an exact digital twin of life on Earth, or of our body, or of our health. And we need, therefore, to expect uncertainty. ”

“The transition to open science and a full, free and open data policy have spurred many digital twin initiatives and are vital for such models. ”

“Models are useful as long as there is a literacy in the communities to translate their results into policies. Otherwise, even the best models will remain useless. ”

“It is critical to realise that scientists are not going to be able to model everything, or even be able to draw definitive conclusions. ”

“A digital avatar project in French Polynesia, rooted in open science, was aimed at helping local governments better prepare, respond, and build climate resilient communities. Such projects use a collective intelligence infrastructure to possibly spur democratic ecological action. ”

“An observatory could be put in place to: 1) capture existing initiatives of “digital twins”, 2) include some oversight in the process to increase trust and 3) ensure citizen engagement, through a digital agora. ”

More information