Website carbon emissions.

The carbon emissions of the internet is an important subject which needs collective action. For those organisations working towards Net Zero, it should be on your action list. Some websites calculate and quote their carbon emissions in grammes of CO2e which seems like a good move forward. However, grammes of CO2e are very small quantities which by nature, implies they are accurate (10 grammes of CO2e is assumed to be far more accurate than 10 tonnes of CO2e). The underlying problem is that the emissions from websites are complex.

If a tree falls in a forest and no one is around to hear it, does it make a sound?

While a poignant philosophical question, of course the answer to this question is yes, the tree does of course still make a sound. The equivalent question is “if a website exists, but nobody looks at it, does it emit carbon?” The answer to this question is also yes. This is because the website lives somewhere on a server (or multiple servers) which all require energy to run, cooling for the data center, maintenance etc. Therefore, the carbon emissions of a website cannot purely be related to the volume of data being transferred to a web browser.

As an example, let’s assume a web server stores a single 1mb image. Over a fixed period of 1 day, the web servers CO2e emissions are 1Kg and it only serves the image to 1 user. That 1 user is therefore responsible for the full 1Kg of CO2e since they were the only person who viewed it. The next day, the server still has the same 1Mb image and the same CO2e emissions, but 100 users view it. Did the server now emit 100Kg of CO2e? No, the server still emitted 1Kg of CO2e but the carbon responsibility of each of the 100 users loading the image is now down from 1Kg (for 1 view) to just 10 grammes per view for 100 users.

Understanding the true factors of website carbon emissions

The true answer for the actual emissions of a website is unfortunately far more complicated than a single CO2e per megabyte figure. To truly calculate the carbon emissions of webpages, it would need a variety of factors to be considered such as:

• The embodied emissions used to create the page in the first instance.
• The emissions of the web server(s) and the data centre which serve the page.
• The number of other sites and the relative volume of data stored on the server which share the storage capacity of the server and are therefore responsible for its emissions.
• The relative proportion of the each of the site’s resources used on the server.
• The distance travelled through the internet from the source server to the user.
• The type of device used to display the page e.g. desktop, lap, mobile.
• The amount of time the user was viewing the page.

A basic example with three variables

To demonstrate just how wildly different the emissions of a webpage can be, we’ll use two scenarios:

User A

• Lives in New Zealand.
• Visits a website from their mobile phone.
• The web server is in France.

User B

• Lives in South Africa.
• Visits the same website on a desktop PC.
• The web server is in Australia.

In these two scenarios, even though both users viewed the same page, User B would have likely generated over 100 times more carbon emissions than User A. This is without even considering all the other factors involved and only based on the three variables mentioned above.

The need for accuracy in website carbon emissions.

For any organisation working towards Net Zero, it is not acceptable to make estimated guesses at its carbon emissions when the actual figures could be so wildly different. Since we cannot be accurate with our website, it would be highly misleading for us to quote a CO2e value. If you would like to make a difference by driving down your website carbon emissions, then visit our low carbon web development page or contact us to talk it through with our team.