Last week I wrote about how the estimated total carbon emissions of the internet have fallen according to the latest Sustainable Web Design Model. This week I’d like to shift the focus from emissions to the mineral resources required to keep the internet running. It’s important not to focus only on carbon emissions. Readers may note that the irony of this being written by the creators of Website Carbon is not lost on the author.
Today I’d like to discuss the concept of carrying capacity and its relationship with our digital lives.
Decarbonisation doesn’t solve everything
The inspiration for this article comes from this report – The environmental sustainability of digital content consumption. It’s a chunky investigation into, well, exactly what its title suggests. According to the report 60% of the world’s population has internet access. Of that 60% the average user spends 40% of their waking life online. If we are spending around 7 hours a day online (2 hours of which are on social media), this is going to have an environmental impact. It is also worth noting that global data traffic levels have risen 440% since 2019. This trend looks set to increase as AI continues to be pushed by all major players online.
The impact of our online activities is most often discussed in terms of carbon output. It’s the simplest way of measuring the impact of an hour’s streaming or loading a webpage. The report looks at the impact of our digital habits on a per capita carrying capacity basis.
Carrying capacity is how many full shopping bags you can carry from the car to the front door:
WikipediaThe carrying capacity of an environment is the maximum population size of a biological species that can be sustained by that specific environment, given the food, habitat, water, and other resources available.
In carbon terms our current online habits account for 40% of our per capita carbon budget consistent with limiting global warming to 1.5 °C. Great news! If we decarbonise the grid, we solve the issue.
As always, it’s not as simple as all that. Focussing on the carbon outputs ignores the pressure on the mineral resources required to keep us online.
Resources and carrying capacity
While our current browsing habits account for less than half of our per capita carbon allowance, that’s not the case for metal and mineral resources. The resources required for the hosting infrastructure, delivery network and devices account for a big chunk of our personal allowance.
Internet usage currently accounts for around 55% of the per capita carrying capacity for mineral and metal resources use. Put simply, rapid decarbonisation of the net can’t mitigate the environmental impacts associated with the minerals required for electronic devices.
The extraction and processing of raw materials have huge impacts. For this article I’ll put the challenges facing the mining and processing industries to one side. It’s a thorny issue but things can be done to improve standards in the industry. One thing that is worth noting is that the impacts associated are regional. Land use, ecotoxicity and working standard issues often occur in locations far from where digital content consumption is taking place.
The % of the connected population will only continue to grow. Increased living standards (and associated improved levels of online access) should be something we strive for. But we can’t ignore the fact that the increase in internet users will also require more devices. You can’t have more devices without more mineral resources. Circular models and recycling won’t solve this issue.
Ideas and solutions
There are technical and practical solutions available to us here. Ultimately though this report leaves us with a more philosophical question around how and why consume digital content.
From a practical point of view, lengthening the lifespan of devices is the first step to take. Using devices and infrastructure for longer has to be key. Choosing refurbished or second hand tech, supporting brands that promote repairability and extending device lifetimes are all practical steps anyone can take. A brand such as Fairphone is a shining example of this ethos. Second hand platforms are widely available but I would advocate for searching for a more local solution if appropriate.
Designing sites and applications that are widely supported and have backwards compatibility for older devices and browsers could help reduce devices obsolescence. Performant and lightweight digital products will work better on older devices too.
As an industry there has to be more scrutiny over the resources used in the AI gold rush. Processing chips are often used once for training large language models and then discarded. Individuals and organisations should carefully question whether the use and support of AI products is worth the tacit approval of the platform’s current working practices.
More than this I think the figures revealed in the report should prompt a rethink of how and why we use digital services. We are back into the digital degrowth territory that we’ve touched on previously.
For organisations and website owners, we should be working towards more optimised user journeys. Less pages and more efficiency equates to less time online and data stored.
We can perform content audits both on sites and social media to identify low performing posts and content. Deleting this information and being more mindful around how much and how often we create and post aids degrowth too.
As individuals, I think it comes down to just how much time we spend online! As a starting point I would ask:
Ultimately all the insights into material usage are more proof that we have a long road to walk when it comes to digital sustainability.
I’d love to hear your thoughts on this article – [email protected]
How can we reduce our digital carbon and resource footprints even as more connected users come online?
How do you feel about your digital consumption habits taking up so much of your mineral resource “quota”?
Andy
Article citation:
Istrate, R., Tulus, V., Grass, R.N. et al. The environmental sustainability of digital content consumption. Nat Commun 15, 3724 (2024). https://doi.org/10.1038/s41467-024-47621-w