Can AI help us tackle climate change?

As nations worldwide clamber to introduce technological solutions to minimise COVID-19’s contagion rate post-lockdown (NHSX Tracking app, remote contact tracing, etc), it’s easy to lose sight of other major developments being made in tackling other crises.

 

Climate change, in particular, should not be forgotten about amongst this global pandemic. Whilst lockdown has admittedly forced us to change our lives in a way that’s ironically much more sustainable, countries are already re-opening workplaces and gearing up their powerplants to once again run at maximum capacity.

 

Ensuring that our post COVID-19 economic recovery is environmentally responsible will be an incredibly difficult task. Global leaders will face pushback to new reforms as the private sector desperately tries to recover its lockdown losses.

 

However, new technologies, or even new applications of existing technology, can make this post-recovery push for sustainability much more achievable. Artificial intelligence (AI) and machine learning (ML) algorithms have made giant leaps in recent years that would have been unimaginable decades ago – and I don’t believe we have even scratched the surface with regards to their potential impact in the battle against climate change.

 

To illustrate this point, below I’ve listed two prime examples of how AI is already spearheading our efforts.

 

It’s snow joke

 

Over the past decade we’ve already witnessed the devastating effects of climate change through unprecedented levels of extreme weather.

 

As sea levels rise and the atmosphere changes, scientists are certain that we’ll soon be experiencing even more intense storms, floods, and destructively strong winds in the years to come.

 

However, a research project combining the efforts of Microsoft, the Montreal Institute for Learning Algorithms, and Conscient AI may equip us with the right tools to adequately prepare for such natural events.

 

Scientists have been gathering swathes of information from every conceivable data point in nature for years; be it ocean temperatures, wind speeds, or soil quality. Such details are useful for studying the progress of the Earth’s global warming – but outside of general trends, the sheer amount of data means that any useful, accurate predictions are extremely hard to make.

 

This aforementioned project, however, uses AI’s ability to sift through massive piles of data and unearth valuable insights at inhuman speeds to do just that. By simulating how entire cities will hold up in any given extreme weather condition – right down to the individual homes – it can show us how the smallest city planner or ecosystem preservation decision can hugely exacerbate problems when disaster eventually strikes.

 

Such exercises are infinitely useful for those who need to plan for such eventualities, such as emergency services. It also provides the most vivid illustration of climate change conceivable to the general public, hopefully spurring greater activism at a local level.

 

Aggressive mist demeanour

 

Preparing for man-made natural disasters is an unenviable, important task. But how can AI be used to ensure that emissions are kept low enough to minimize the chances of such catastrophe ever occurring?

 

The answer comes from the independent think tank Carbon Tracker and their image analysis machine learning (ML) algorithmic programs.

 

A state or international agreement can try and enforce any number of pollution limits or carbon tax thresholds for industrialists to follow. However, actually ensuring that they do necessitates frequent inspections, though even those are impossible to implement in more reclusive nations.

 

Carbon Tracker have, in effect, somewhat solved this problem. By feeding satellite imagery of powerplants across the world through their AI-powered algorithms, they can quantify the levels of emissions from any polluting site on earth.

 

It’s impossible to imagine a human being able to measure levels of pollution purely from images taken 36,000 kilometres in the air, but for AI it is an extremely achievable task. From analysing millions of images of powerplants and combining them with the accurate data of previous emissions, it can then cross-reference such information – in real time – to give activists and researchers an incredibly powerful tool for identifying the worst offenders.

 

Such a remarkable technology can easily identify those polluting to illegal levels and provides a fantastic disincentive for any other potential polluters to try and meddle with their own figures. A recently gifted grant from Google will undoubtedly help refine this process and, eventually, help release it into the world – allowing everyone to hold those harming our planet to account.

 

These examples represent only a small sub-section of the incredible ways AI is being used in the war against climate change. Sustainability technology has already made fantastic leaps in recent years and, given the promise shown in these two examples, I don’t doubt that we will be seeing even grander innovations made in the near future. The better prepared and equipped we are to save our planet, the greater chance we have of pulling it off.

 

 

Nikolas Kairinos is the chief executive officer and founder of Fountech.ai, an umbrella company to three specialist firms: Fountech.Solutions, Fountech.Ventures and Fountech.Science. Fountech.ai is driving innovation in the AI sector, helping consumers, businesses and governments understand how this technology is making the world a better place.