Last week I appeared on Kay Burley’s Breakfast Show on Sky News to discuss new stats about the carbon footprint of “unwanted” photos.
According to the press release from The Institution of Engineering and Technology (IET):
New research today reveals Brits’ hidden dirty data habits, with the nation’s trigger-happy social snappers contributing over 355,000 tonnes of CO2 every year through unwanted pics alone: the equivalent to the entire population of Chelmsford flying to Australia and back. The new study by the Institution of Engineering and Technology (IET), found just a quarter of respondents delete additional shots they take, leaving millions of identical images being added to storage every week. And for those that do delete their excess pictures, fewer than one in six (16%) say they do this for environmental reasons (i.e., to reduce the burden of energy needed to power servers used to store our data dumps). With the average person taking almost 900 photos per year the duplicated, unwanted images left in storage alone could accumulate 10.6kg of CO2 emissions annually for every adult in the UK – the equivalent of over 112,500 return flights from London to Perth, Australia.
As I said on-air, there are problems with these figures. In this post I’ll discuss why I think they’re bogus.
Photos on the phone
Thanks to regulation (US and EU), electrical devices have become more and more efficient over the last few decades. This means the majority of the carbon footprint of the phone is in the manufacturing stage. For example, the latest Apple iPhone 13 is responsible for 64 kg of lifecycle carbon emissions, only 16% of which is its usage (over an assumed 3 year lifespan).
Phones have an internal battery which is charged periodically, generally overnight. Battery drain depends on the usage pattern, and mobile operating systems have been getting more sophisticated power management over the years.
The file format also has an impact. In 2017, Apple introduced a new file format – HEIF – which uses up to half the storage size as an equivalent JPEG. Other operating systems followed. Is this factored into the calculations?
Calculating how much energy is consumed by taking a single photo is difficult. You’d need to measure the tiny amount of power consumption before and after taking a photo. Accurately inspecting that energy consumption is challenging because of factors like other operating system tasks, the charge state of the battery, and the ability to measure such a small unit of work. Taking a photo takes less than a second. Compare that to 1 hour of video streaming, which gives a much larger period to measure over.
Taking the photo is just one component. Storage is another. All phones use solid state disks which have a set power consumption during read/write, but have a very low profile when idle.
Then you’d need to multiply by the carbon factor for the power grid where the phone was charged.
Phones are charged by electricity, which has a carbon intensity factor that will depend on the grid mix when it was charged. As this “study” was focused on the UK, we can say that over the past 10 years the UK grid mix has shifted significantly towards renewable electricity. There is a lot more progress needed, especially internationally, but the electricity grid is rapidly decarbonising.
The study doesn’t provide any reference year, so are they refering to the carbon emissions in 2009…or 2018? The carbon factor for the electricity used to charge the phone would be very different in each case.
Either way, the grid emissions are not something a phone user has much control over, and will change every time the phone is charged.
Photos in the cloud
Once taken, the photo might then be uploaded to the cloud. If you’re on an Apple device this will probably be iCloud. On Android, it’ll probably be Google Photos. It may also be shared on Facebook (via Instagram, WhatsApp or Facebook itself). Depending on the settings, the photo may then be automatically deleted from the device (usually to save space).
When the photos are uploaded also has an impact. The default settings usually mean photos will only be uploaded when the phone is on wifi (which is less energy intensive than cellular) and when it has sufficient battery charge.
In the case of iCloud, Apple has been running it’s data centres on 100% renewable electricity since 2013 (and all its global facilities since 2018). Google has been doing this since 2017. Facebook is not quite as good, but achieved this as of 2020. Although that doesn’t actually mean the electricity going into the data centre has been generated from 100% renewable sources 24/7, the combination of renewables matching with offsets means that on a reporting basis there is zero carbon associated with storing these photos in the cloud.
We also don’t know the architecture of the systems behind these cloud services. If you have ever used a cloud storage product like AWS S3 or Google Cloud Storage, you’ll be aware of the different storage tiers you can set up depending on how frequently you access the data. Photos just uploaded may be stored in hot storage so they can be viewed/shared immediately, but over time they may be moved into cold storage for infrequent access.
For example, Facebook has written about this in the past:
As the number of photos continued to grow each month, we saw an opportunity to achieve significant efficiencies in how we store and serve this content and decided to run with it. The goal was to make sure your #tbt photos from years past were just as accessible as the latest popular cat meme but took up less storage space and used less power. The older, and thus less popular, photos could be stored with a lower replication factor but only if we were able to keep an additional, highly durable copy somewhere else.
Under the hood: Facebook’s cold storage system
The network carbon intensity is not factored into this – and individual ISPs still have a lot of work to do compared to the big cloud providers – but placing photos in the cloud is the most sustainable location for them.
Where are the emissions?
The number cited by IET is 10.6kg of annual CO2 emissions. This is not from the 900 photos that each adult supposedly takes each year, but an “unwanted” subset of that total.
This is based on an original survey carried out by the IET which says that 25% of respondents delete additional shots they take, and 69% do not delete duplicated photos.
What does “unwanted” or “additional” mean? It’s not defined.
So if we assume that 25% of 900 = 225 “unwanted” photos, the IET are saying that 225 photos are responsible for 10.6kg of emissions per year. 10.6 / 225 = 0.047 kg per photo, or 42.4kg for 900 photos.
How realistic is this? We can check it against the lifecycle emissions of an iPhone 13 = 16% of 64kg = 10.24kg. But that’s over a 3 year lifecycle. That means 3.41kg per year of full use.
These numbers don’t make sense – the IET number is x12 larger! Apple reports that using the phone for a full year will cause 3.41 kg of carbon emissions. The IET says that taking 900 photos per year will cause 42.4kg of carbon emissions.
If there are zero emissions associated with cloud storage, where does that figure come from? It’s clearly bogus.
Poor quality research
This release by the IET is an example of poor quality research with flawed assumptions designed as click-bait to get news coverage timed to land during COP26. It has multiple problems:
- There are no workings. The sources for the numbers are not provided, and the references do not back up the claims.
- Simplistic calculations. There is no understanding of the lifecycle emissions of the phone, the varying grid mix is not accounted for, where the photos are stored is not considered, the emissions profile of the cloud is ignored, the storage architecture of the cloud is not discussed.
- There is no sense-checking against readily available figures. A figure that is x12 higher than the lifecycle analysis published by the manufacturer doesn’t make sense.
- The conclusions are based on a survey of user behaviour that uses vague terms to model “bad” behaviour. Who decides what is “unwanted”? That will be different for every person.
- It ignores the published research showing that energy consumption has decoupled from demand.
- These conclusions are used to recommend that deleting a few photos or clearing your WhatsApp chats will have a meaningful impact on the environment. This makes people feel good without having any actual impact.
The way to make meaningful change is to pressure organisations to decarbonise their supply chains and infrastructure. When Google switches its data centres to renewable electricity, users get that benefit by default. When Apple improves the environmental status of its supply chain, users get that benefit by default.
Trying to convince individuals to change their behaviour is the wrong place to focus. A few people might adjust their habits, but the global impact rounds to zero. It’s significantly easier to decarbonise electrons than atoms. If you want to have an actual impact on your carbon footprint, not buying a new phone or laptop is the best way to do so.
It’s disappointing that the IET decided to waste time on these absurd scare figures when they could have had much more impact by highlighting how little we know about most supply chains. Google and Apple are in the minority of companies making progress. If they want to focus on tech, there’s plenty to criticise (water consumption, for example). We know almost nothing about other industries. That’s why transparency is the first step to catalysing change.